Skip to main content

Full text of "Lectures on the history of Rome, from the earliest times to the fall of the Western Empire. Eidted by Leonhard Schmitz. 2. ed., with every addition derivable from Isler's German ed"

See other formats

-?^(c - '?7; 

.^r.- 377- 172 






1: ^ 


Dr. R. G. Latham's English Grammar, 

For the Use of Schools. Fourth Edition. 12mo. 4s. 6d. cloth. 

TifK ohjett of this work is to lay down the principles of English Grammar as 
distinguished from mere iiiles. The student in the middle and higher classes of 
schools, for wiiom the work is chiefly intended, is made acquainted with the con- 
nexion of the varif)us liranches of the great Gothic stock of languages; and, by 
learning the history of liis own tongue, and its relations to the dialects with which 
it is coimected, acquires the elements of the general })hilological dassiticatifin of 
languages. The book presents him with new and numerous facts, and habituates 
him to reasoning upon them ; and while the work can be wholly mastered, 
independently of any knowledge of either of the classical languages, as much 
logic is given as is necessary to explain the structure of propositions. 

Dr. R. G. Latham's English Language. 

Third Edition. 8vo. ' 15s cloth. 

In this work, the Histor}' of the English Language is traced from its remote 
origin, through its successive changes and periods, to its present state. The nature 
of its connexion with all the languages, which either form its basis or have been 
in any degree incorporated with it, is minfitely detailed. The historical 
portion is everywhere illustrated by extracts from primitive books, records, and 
inscriptions ; by analogies drawn from the Sanskrit, and classical languages, and 
from the Gothic, Celtic, and Sclavonic dialects of ancient and modern Europe ; 
and by comparative catalogues of derivations, affinities-, and provincialisms. 

Dr. R. G. Latham's Elements of English Grammar, 

For the Use of Ladies' Schools. Fcap. 8vo. Is. 6d. cloth. 

'' Instead of a mere diy classification of words and rides, often difficult to be 
understood, it gives a philosfiphical view of the whole subject, and yet with so 
much clearness as to assist the reason no less than the memory. The lirst part 
gives the history of our language and traces its formation from the German 
tribes who successively conquered and settled in our island. In the third part, 
which treats of declension and conjugation, the connexion of these subjects with 
the rules of euphony, is shown in a very lu<id and interesting; manner. We think 
this little work will prove a boon to mothers and instructors, and invest with 
interest a branch of education which has often proved one of the most uninviting- 
steps in the ladder of learning." — Qu<,rlir'y Ediaationut Mayazine, No. VIII. 

History and Etymology of the English Language, 

For the Use of Classical Schools. By Dr. R. G-. Latham, F.R.S. 
Fcap. 8vo. Is. Grf. cloth. 

This is an introductory grammar, intended for those desirous of studying the 
English language with I'eference to the history, the transitions, and the analogies 
of its forms. 

Elements of English Grammar, 

For Commercial Schools. By Dr. R. G. Latham, F.R.S. Fcap. 

8vo. Is. Qd. cloth. 

The first part of this grammar contains, besides a concise history of the 
origin and changes of the lang-uage, an enumeration of the places where it is 
at present universally or partially spoken. The section on Sounds and Letitr* 
includes a condensation of the treatise prelixed to Walker's Dictionary, and 
settles the promuuiation of many doul)tful words by a reference to ttie best 
mod(!rn usage. The fundamental principles of the language are elucidated in tl.e 
sections on Injiectiun and Derivatiun, and on Syntax. 

Latham's First Outlines of Logic, 
Applied to Grammar and Etymology. 12mo, Is. 6d. cloth. 


^bz/^^/ ^hCT^ crU.^ ^ /^- ^"trr-i^nz^ 





ci.s>'-" : 




the earliest tbies to the fall of the 
^\t:stern empire. 





ftcronti iStiition. 


VOL. I. 











>/^ ^ 


When In 1844, I published tlie Lectures of Niebuhr, 
embracing the History of Rome from the commencement 
of the first Punic war down to the death of Constantine, I 
entertained a strong hope that Niebuhr's friends in Germany 
would be roused to a sense of duty, and no longer with- 
hold from the world his valuable relics in his own language. 
In that hope I was, for a time, disappointed; for no sooner 
were the Lectures published in this country than there 
appeared, at once, advertisements of two German transla- 
tions of them. The idea of translating from English into 
German a work of which there existed in Germany numerous 
manuscripts containing the very words and expressions of 
Niebuhr, and which required only the carcfid and conscien- 
tious supervision of an editor, seemed to be a somewhat 
preposterous undertaking. If the Lectures were to be 
published in Germany, assuredly the German public had a 
right to expect that the exact language of the historian should 
be scrupulously preserved, which is an impossibility in a re- 
translation, in the execution of which, moreover, no use was 
to be made of the manuscript notes taken by the students 


during the delivery of the Lectures. Only one of the adver- 
tised translations, however, made its appearance; and that 
was more than enough, for it bore so many marks of careless- 
ness, and displayed so flagrant a want of knowledge of the 
English language, that even the most moderate expectations 
were disappointed. As there was reason for believing that 
every succeeding volume of Niebuhr's Lectures which might 
appear in this country would meet with the same fate in 
Germany as the first two, and that an unpardonable wrong 
would thus be done to the memory of the author, M. Marcus 
Nicbuhr, the son of the historian, and some of the more 
intimate friends and pupils of Niebuhr issued an announce- 
ment, that they would forthwith set about preparing 
a German edition of all Kiebuhr's Lectures, on the only 
principle that could secure for his memory that honour among 
his own countrymen to which he is so justly entitled. Thus 
the very circumstance which at first had seemed to thwart 
my hopes contributed in reality to their speedy realisation. 

The task of preparing the German Edition was undertaken 
by M. Marcus Niebuhr, Dr. Islcr of Hamburgh, and Professor 
Classen of Llibeck. My co-operation also was solicited ; 
but other engagements prevented my accepting the honour- 
able proposal ; and it was finally arranged that I should under- 
take the Editorship in England of the whole Series of 
Lectures. The first volume, containinsc the Lectures on the 
History of Home from the earliest times down to the com- 
mencement of the first Punic war, edited by Dr. Isler, 
appeared at Berlin in 1846. Of this a translation is now pre- 
sented to the English public. As to tlie materials of which 
the German editor has made use, and the plan he has followed, 
I shall do best to let him speak for himself " The History 
of the Roman Republic," lie says, " is one of those few subjects 


on which Nicbuhr gave two courses of Lectures in the 
University of Bonn; the first in the winter of 1826-7, and 
the second during the winter of 1828-9- In the summer of 
1829, he lectured on the history of the Roman Emperors 
down to the overthrow of the Western Empire. In the course 
of 1826, he did not carry the History further than to the 
time of Sulla ; but in many parts of it he entered more minutely 
into the criticism and analysis of the existing materials; and 
this circumstance prevented him from carrying the History as 
fur down as in the latter course of 1828. "V\Tiat is here 
presented to the reader, consists essentially of the latter course 
of Lectures; but all that is of interest or importance in the 
earlier one of 1826 has been incorporated, wherever it seemed 
appropriate. This combination of the two courses of Lectures 
into one, though it does not always pi-eserve the exact form 
and order in which Xiebuhr related the History, yet does not 
contain a single idea, nay hardly a single word, which Avas not 
actually uttered by him. If this should be thought an arbitrary 
mode of proceeding, the editor takes the responsibility upon 
himself; but he must at the same time state, that he con- 
sidered this to be the way in which the treasures entrusted to 
his care could be disposed of in the most careful and con- 
scientious manner. A considerable number of manuscripts 
have been collated, and all the available materials have been 
scrupulously sifted and weighed, in order to ensure the value 
of the work as much as possible. The editor's labour has been 
of a purely philological nature, inasmuch as it was necessary 
to form, as fiir as it could be done, a genuine text out of a mass 
of notes presenting such discrepancies and inaccuracies as natu- 
rally occur in notes hurriedly made by students in the lecture- 
room. Those who are acquainted with such matters know 
that the formation of the text consists not only in restoring 



the exact expressions of the Lecturer, but also in tracing the 
facts stated to their respective authorities, wherever practi- 

Dr. Isler further states, that when his manuscript was ready 
for the press, it was revised by Professor Classen ; and that 
M. I\Iarcu3 Niebuhr, who afterwards undertook the revision of 
the proof-sheets, also suggested several improvements. From 
these statements, the reader will see that the German editor 
had greater advantages than could have fallen to the lot of 
any one undertaking the task in this country ; for he not only 
had notes from two distinct courses of Lectures on the same sub- 
ject, one of which was supplementary to the other, but he was 
assisted by those most deeply interested in the work. As, 
moreover, the three volumes of Niebuhr's immortal History 
treat of the same period as that contained in these Lectures, 
the former always served as a corrective, wherever the manu- 
scripts of the latter were obscure or imperfect. 

Under these circumstances, I might have confined myself 
to the mere translation of the present Lectures; but as I 
possessed some very excellent manuscripts, I thought it right 
to institute a careful collation of them; and my labour has 
been amply rewarded, for I found a considerable nimiber of 
most interesting remarks and statements which do not occur 
in the German edition, so that in many respects the present 
volume is more complete and perfect than the work on which 
it is founded. Dr. Isler has not di\'ided his edition into Lec- 
tures, because the Lectures in the two courses did not always 
correspond, or treat of the same subject; but in the present 
work the Lectures have been kept distinct, partly because I 
consider that division to be essential to a right understanding 
of the work, and partly for the sake of consistency, the same 
plan having been adopted in the two volumes published in 


1844. In doing this, however, I was under the necessity of 
making some Lectures disproportionately long, as passages of 
considerable extent or even entire Lectures from the course 
of 1826 had to be inserted in Lectures of the course of 1828; 
while, by the transfcrcnee of passages from one Lecture to a 
more appropriate place in another, some Lectures will appear 
rather short. 

It may perhaps be asked, What is the use of publishing the 
Lectures on that portion of Koman History on which we 
possess the author's own elaborate volumes? To this it may 
be replied, that the present Lectures contain a more popular 
and familiar exposition of the subject, which in the three 
volumes is treated in a severe style, little calculated to attract 
ordinary readers. They, therefore, may be used as an intro- 
duction to, or as a running commentary on, Nicbuhr's great 
work. I also agree with the German editor in thinking that 
it does not seem right to suppress any part of the Lectures on 
Roman History, one of the objects of their publication being 
to give as vivid a picture as possible of the extraordinary 
personal and intellectual character of Niebuhr; an object 
which can be attained only by the complete and entire publi- 
cation of all that he has ever said on the history of Rome. 
These Lectures, moreover, as Dr. Isler remarks, " distinctly 
show the different objects which Niebuhr had in view 
in preparing a work for the press, and in lecturing 
from the professorial chair; each, in his opinion, demanded a 
totally different mode of treatment, whence many points are 
set forth in these Lectures more clearly and distinctly, nay 
sometimes even more minutely than in the larger work. The 
reader need only be reminded of the Introductory Lectures on 
the Sources of Roman History, of the Discussion on the Sa- 
turnian verse, and the like. Lastly, it must not be forgotten, 


that on many subjects these Lectures contain the latest and 
most matured opinions of Nicbuhr. The revision of the 
lust edition of the first volume of his History was finished by 
him, chiefly, in the year 1826; and the additions to the third 
edition belong to the year 1827. A mind like that of 
Nicbuhr never ceased acquiring fresh stores of knowledge, and 
making new inquiries, although the principal results were 
already firmly established. Sundry new fragments of ancient 
writers also were discovered after the publication of the last 
edition, which led him to modify the views he had exjjressed 
in his printed work. In regard to the period treated of in 
the third volume, the reader will find in these Lectures many 
additions and corrections ; for the greater part of that volume 
was composed as early as 1812, and if Niebuhr had lived to 
prepare a new edition of it, he would undoubtedly have intro- 
duced many important alterations. Hence even those who by 
a careful study have acquired a thorough familiarity with the 
three volumes of the Koman History, will find in these Lec- 
tures much that is new and strikino-." 



Edinburgh, Nuvemhe7\ 1847. 



Origin of Roman History 1 

Impossibility of the earliest history 1 

Numerical schemes in the chronological statements 3 

Saecula of the Etruscans 4 

Ancient Lays 5 

Etruscan History , 7 

The emperor Claudius 7 

Saturnian verse 9 

Neniae 10 

Ej)ic poems 11 

Family records and family vanity 12 

National vanity 12 

The Pelasgians 14 

Samothracc 15 

Siculians, Itali, Ocnotrians, Peucetians, Liburnians, Tyirlicnians, Opicans, 

Apulians, Volscians, Aeqiiians, Sabcllians, Umbrians 16 


Siculians in Italy, Aborigines 19 

Latins 19 

Polarity of Traditions 20 

Cascans 21 

Sacrani, Ver Sacnun, Prisci, Prisci Latini 22 

Origin of the Latin language 22 

Tradition about the Trojan settlement in Latium 23 

Alban Chronology 34 

Alba Longa; populi Albcnses 24 


The thirty Latin towns 26 

Roma, the town on the Palatine 27 

Romulus and his descent 28 

Romidus and Remus 37 

Rcmuria 37 



The Asvlum 33 

Rape of the Sabine women 33 

Union of the Romans and Sahines ,, 34 

Death of Romulus 35 

Di\'ision of the population 35 

The Sahincs 36 

Towns on the Palatine and Quirinal 37 

Union of the two states 37 


Division of the population 38 

The Sabines and the Palatine and Quirinal 39 

Double state 39 

Numa Pompilius 41 

Tullus Hostilius 41 

War with Alba 42 

The third tribe 46 

Ancus Marcius 46 

War with the Latins 46 

Foundation of Ostia 47 

Origin of the Plebs 48 


Tarquinius Priscus and his Greek descent 48 

The Cloaca maxima 53 

Ti'aees of Rome ha\'ing then been a great state 54 

The centuries doubled 55 

The Etruscans and T^TThenians 57 

Sen-ius Tidlius (Mastama) 67 

Constitution of Sendus Tullius 69 

Gentes and Curiae 71 


The plebes or commonalty, and the clients 74 


The plebeian tribes , 82 

The centuries 85 


Centuries continued 86 

Census 90 

Further legislation of Servius Tullius 93 

Relation to the Latins 94 

Extension of the city 96 

The cloacae 97 

Mound of Servius Tullius 98 


Criticism on the tradition about Mastarna 99 

L. Tarquinius Supcrbus . 102 


LECTURE X. ^.^„ 


War with the Latins 103 

Treaty with Carthage 104 

Military constitution 105 

L. Junius Bmtus 106 

Abolition of the kingly government 1 09 


The consulship 110 

Valerius Poplicola and the Valerian laws 113 


Porsena 114 

War of the Etruscans against Rome 115 

Mucins Scaevola 117 

Peace of Porsena 119 

Diminution of the tribes 121 


Relation of the Latins 121 

Battle of lake Regillus 122 

Isopolity 124 

Treaty of Sp. Cassius, league of the Romans, Latins and Hemicans 125 

The dictatorship 125 

War with the Auruncans 126 


Attempts to bring about a counter-revolution 128 

The law of debt 130 

The nexum 133 


Resistance of the plebes and their secession 135 


Peace between the two estates 141 

Tribunes of the plebes 143 

The story of Coriolanus inserted in a wrong place 146 


Division of the Volscian wars 148 

League with the Hernicans 149 

Sp. Cassius 150 

The agrarian law 152 


The agi-arian law continued 154 

Dincrcnce between property and possession 156 

Tiie lex Cassia 157 

Execution of Sp. Cassius 15S 



The consuls clcctcil liy the senate and curios alone 1 60 

The election of the consuls divideil between cui-ies and centui-ics 161 

War against Vcii 1 62 

The Fahii declare for the plebeians 162 

The Fabii on the Crcmera, and their desti-uction 163 

Accusation of the consuls by the tribunes 165 


Murder of Cn. Gcnucius 167 

Volero Publilins 167 

The PuUilian rogations 168 

Mode of i)roceeding in the assemblies of the people 169 


Resistance of Appius Claudius 171 

Wars with the Volscians and Aequians 1 73 

Plague at Rome 176 


C. Tcrentilius Harsa and the TerentUian law 177 

Kaeso Qiunctius 180 

Cincinnatus 181 

Appius Herdonius 182 

Condemnation of Volscius 183 


Coriolanus 184 

Peace Avith the Volscians 190 


Altered Relation of the Latins to Rome 191 

Commotions at Rome 193 

P. Mucius 193 

Embassy to Athens 194 

Ilcnnodonis 195 

First dccemvirate 196 

Equalisation of the rights of patricians and plebeians 197 

Second dccemvirate and new constitution 197 


Unlimited right to make a will 198 

The law of debt 201 

TIic centuries, a national court of justice -02 

Tyranny of the decemvirs 204 

Death of Virginia 208 

Secession of the plebes 208 

Overthrow of the dccemvirate 209 


Restoration of the old constitution 209 



Veto of the tribunes, and patrician tribunes 210 

Death of App. Claudius and Sp. Oppius 212 

The Roman criminal law 213 

Lex Iloratia Valeria 216 

Ilortensiau Law 217 

Victories over the Aequiaus and Saltines 218 

The different quaestors 219 

The couuubium between the two estates, and the Canuleian law 220 


The military tribunes 222 

The censorship 226 


Famine at Rome, Sp. Maelins 230 

Tlie executive power of the consuls , 232 

The quaestorship openod to the plebeians, aud plebeian senators 232 

The Campauians 234 

Vietoiy over the Aequiaus 230 

The agrarian law 237 

Coloniae Romaniae 238 

InsuiTCction of the soldiers 238 


Destruction of Fidcnac 240 

Mihtary aH'airs and pay of the army 241 

Siege of Veil 243 

The Alban lake and its emissariuni 24.'j 


The taking of Veii and the disputes between patricians and plebeians 

resulting from it S.'JO 

War with the Faliscans and Vulsinians 251 

Camillus and his exile 252 

The migration of the Gauls 253 

The Celts 255 


The Celts 256 

The Gauls invade Italy 259 

Emba.ssy to the Gauls 261 

Battle of the Alia 262 

The Gauls in Rome 205 


Peace ^vith the Gauls, and their departure 268 

Consequences of the Gallic conquest 272 

Rebuilding of the city 273 

Foenus uueiarium 275 

Usury 275 




Etruscan Wars 276 

Four new tribes 276 

M. Manlius 279 

Tribuneship of C. Licinius Stolo and_L. Sextius Lateranus 282 

The Liciuian rogations 283 


Dictatorship of Camillus 285 

Temple of Concord 289 

The Consulship divided between patricians and plebeians 289 

The praetorship 290 

Ludi Eomani 290 

The curule aediles 291 


Triumviri reipublicae constituendae 292 

Invasion of the Senonian Gauls 293 

Alliance with the Latins and Hernicans 294 

Alliance with the Samnites 296 

War in Etraria 297 

Settlement of debts 298 

Third invasion of the Gauls 298 


C. Marcius Rutilus 299 

The colonies 301 

Origin of the Samnites 302 

InsiuTCCtion at Capua 304 

Constitution of the Samnites 304 

The first Samnite war 306 

M. Valerius Coi'vus 308 

Battle of mount Gaurus '. 309 

P. Decius Mus saves the Roman army 311 


Insun*ection in the Roman aimy 312 

Progress of legislation 314 

Military arrangements of the Romans 315 

Peace with the Samnites 317 

Relation between Rome and Latium 318 

War with the Latins 319 

T. Manlius 320 

The Roman anny 321 

Battle of Vcseris 321 

P. Dcoius 323 


Battle of Trifanum 323 

Submission of tlie Latins 324 

Q. Publilius Philo and his legislation 325 



End of the Latin war 326 

Mimicipia 327 

Latin colonies 330 


War with the Sidicines 334 

Colonies at Cales and Fregellae 334 

Now circumstances of the Romans and their relations to Greece 335 

Tarentum and Alexander of Epirus 337 


InsmTdction of Privernum 343 

Peace with the Gauls 346 

Embassy to Alexander of Macedonia 347 

Palaepolis and Neapolis 348 


Outbreak of the second Samnite war 349 


M. Valerius Con-us, L. Papirius Cursor, and Q. Fabius Maximus 3.')8 

Fabius defeats the Samnites, and escapes from Papirius 3G0 

Death of Papius Brutulus 362 

Defeat of Caudium 364 

The Romans break through the peace 366 


Defeat of the Romans at Lautulae 367 

Colony at Luceria 368 

The Romans build a fleet 372 

Art among the Romans 373 

Rise of the Etiniscans 374 


Taking of Bovianum 374 

Papirius Cursor appointed dictator 375 

War with the Hemicans and their subjugation 376 

End of the second Samnite war 378 

Reduction of the Aequian'? 379 

Rome's connection with the Marsians 379 

The Etruscan war 380 

The Ciminian forest 380 

Battle of Sutrium 387 

Colony at Narnia 383 

Cleonymus 883 


Appius Claudius the Blind 3S4 

Via Appia 389 

Aqiux Appia 3'JO 

Cn. Flavins 39 1 



Abolition of the ncxiim 393 

The Ogulnian law 394 


The third Samnite war 395 

Tlie war transferred to Etniria 396 

Battle of Sentinum 401 

P. Decius devotes him'^clf for his coiiutry 401 


End of the third Samnite war 403 

War with tlie Sabines 405 

Embassy to Epidaiirus 40G 

Draining of lake Velinus 408 

The falls of Terni 408 


The Macnian law 409 

The Hortcnsian law 410 

Triumviri capitales 414 


War against the Scnonian Gauls 416 

C. Fabricius Luscinus and M.' Ciirins Dentatus 417 

Ti. Coruncanius 418 

Outbreak of the war with Tarcntum 419 

PpThus of Epinis 421 


PjTrhus, continued 421 

Cinea.s 424 

Battle of Heraclea 427 

Pyn-lnis attempts to march against Rome 428 

Sends Cineas to Rome 429 


Pyn-lius returns to Tarentum 430 

Roman Emljassy to Pyn-luis 43 1 

Battle of Aseulum 432 

Pyrrhus goes to Sicily, and retm-ns to Italy 434 

Battle of Taurasia (Bcncventum) 435 

Defeat of Pyn-hus 436 


Peace with Samnium 437 

Tarentum taken by the Romans 438 

Suiijugation of Italy 439 





At the time when Fabius began to write the history of Rome, 
his materials consisted of the annales pontificum, the fasti, the 
libri pontificum and augurales, the laudationes, and poetical lays. 
Of the mea^reness of these materials we have already satisfied 
ourselves; but what was their authenticity? They might have 
been not less authentic than our INIcrovingian and other ancient 
annals ; nay, as the annales pontificum began ab initio rerum Eoma- 
narum, or at least from the time of Nunia, theij might have been 
very authentic; in them, as we are informed by Dionysius, 
the pontiffs had recorded with the utmost accuracy every year 
of the kingly period ; and the triumphal fasti even mentioned 
the very days on which the kings hod triumphed over their 

But the consideration that the early history, such as it has 
come down to us, is impossible, must lead us to enquire whether 
the earliest annals are deserving of credit. Our task now is to 
prove that the earliest history does contain impossibilities, that 
it is poetical, that the very portions which are not of a poetical 
nature, are forgeries, and, consequently, that the history must 
be traced back to ancient lays and to a chronology which was 
invented and adapted to these lays at a later period. 

The narrative concerning the primitive times given by Livy 
dilfers considerably from that of Dionysius; Livy wrote his 
first book without assigning the events to their particular 
years, and with an extraordinary want of criticism: he here 
evidently followed Ennius, as we may see by comparing the 
fragments of the poet's writings with tlie statements of Livy; 
compare, for example, Livyii. 10 with the fragment of Ennius: 

VOL. I. B 


Teque pater Tiberine tuo cum numine sancto. Dionysius, at- 
tempting to make out a true history, proceeds on tlie supposition 
that the detail of Roman history can be restored, and that the 
historical ground- work is only overgrown with legendary tales ; 
he endeavours to reconstruct the former in an arbitrary man- 
ner, and inserts his pragmatical speeches in his account of the 
mythical ages, whereby he often makes himself truly ridiculous. 
Livy, on the other hand, wrote the history such as he found 
it in the most ancient books and as it appeared to him the 
most beautiful; he gives it in its ancient form before it was 
artificially corrupted; and hence his narrative is the purest 
source for the history of those times. 

The story of the miraculous conception of Romialus is an 
historical impossibility; although in the school of Piso it was 
metamorphosed into an history : the same must be said of the 
account of the rape of the Sabine women, whose number was 
thirty in the original tradition, and also of the ascension of 
Romulus during an eclipse of the sun.^ Such also is the cha- 
racter of the long reign ofXuma with its uninterrupted peace, 
and of his marriage with the goddess Egeria, which among the 
contemporaries of Scipio was as implicitly believed as the 
history of the Punic wars. The story of the combat of the 
Horatii and the Curiatii, who were born on the same day of 
two sisters has a very ancient poetical character.^ We next 
come to Tarquinius Priscus, who was already married to 
Tanaquil when he migrated to Rome in the eighth year of the 
reign of Ancus (which lasted twenty-three years). Tarquinius 
liimself reigned thirty-eight years and<was at his death upwards 
of eighty years old, leaving behind him children under age 
who were educated during the forty-three years of Servius's 
reign, so that Tarquinius Superbus must have been at least 
fifty years old when he slew his father-in-law. Tanaquil lived 
to see t]iis crime, and required Servius to take an oath not to 
resign his crown: at that time she must have been 115 years 
old. One of the first features in the story of Servius is that on 
one occasion in his infancy his head was encircled with a flame, 
which Dionysius attempts to explain in a natural way. CoUa- 
tinus is said to have been the son of a brother of Tarquinius 

■ The moment at which Mars overcame Hia was likewise marked by an 

eclipse of the sun. — N. 

' Li^-j-'s account is already somewliat disfigiu'cd. — N. 


Piiscus, and this brotlicr, it is stated, Avas born previously to 
the migration of Tarquinius Friscus to Rome, that is, 135 years 
before the expulsion of Tarquinius Supcrbus; and Collatinus 
is described as being a young man thirty years old, at a time 
upwards of 120 years after his father's birth. Brutus is said 
to have been Tribunus celerum, which was the first place in the 
equestrian order, in which he represented the king, assembled 
the senate, and was obliged to perform the most important 
sacrifices; and this place the king is stated to have given to a 
man, whom he thoxight to be an idiot and whom, for this 
reason, he had deprived of the management of his own pro- 
perty ! Brutus, the story goes on to say, feigned idiocy for 
the purpose of escaping the envy and avarice of the king. He 
is described as the son of a daughter of Tarquinius Priscus, 
and as dreading to enrage the king by taking possession of his 
own property: — but Tarquinius did not even belong to the 
same gens. At the beginning of the reign of Tarquinius 
Superbus, Brutus was only a child, and immediately after the 
king's expulsion he appears as the father of sons who have 
attained the age of manhood. 

All these chronological points, to which many others might 
be added, even down to the time of Camillus, bear so much 
the character of absurdity and historical impossibility, that we 
are obviously entitled to criticise. Now let us remember the 
two-fold sources of the earliest history of Eome, namely, the 
chronological: the /o5^i and annales pontificum ; and the un- 
chronoloo'ical : the lays, lauclationes, the libri jwntificum and 
nurjurales. As regards the chronological sources, in the most 
ancient account, that of Fabius, we find 3G0 years reckoned 
from the building of Rome to its destruction by the Gauls, 
exactly the number of the 7eV;; in Attica, which number was 
declared, even by the Greeks, especially by Aristotle, from 
whom the grammarians Pollux, Ilarpocration and others de- 
rived their information, to be that of the days in the solar 
year. J>ut the number 3G0 if accurately examined Avill be 
found to be the mean number between the days of the solar 
and those of the lunar year, and the nearest to each that can 
be conveniently divided. Of this period of 360 years, the 
time assigned to the kings was, according to the earlier calcula- 
tion, 240 years, and that to the republic 120 years. Thisnuiu- 
ber has as much of a mathematical character as that of the 


Indian ages of the world, tlie Babylonian and other Oriental 
numbers. The 120 years assigned to the republic is adopted 
even by those writers who calculate the whole period at 365 
years. Whether 120 years be correct, must be determined 
acccrding to the view respecting the time at which the Capitol 
was consecrated. That the annales pontificum were destroyed 
in the Gallic conflagration is strongly confirmed by Claudius 
(vmdoubtedly Claudius Qviadrigarius) as quoted by Plutarch, 
and indirectly by Livy, who could not state it directly, since 
he would thereby have declared the first books of his own 
work valueless; it is moreover confirmed by the fact, that the 
eclipse of the sun in the year a.u. 350, the first which was 
actually observed, was mentioned in the annals, whereas the 
earlier ones were subsequently calculated, and, as we may safely 
infer, considering the means of the science of that time, were, 
of course, calculated wrongly. For the first 240 years we have 
seven kings, whose reigns are said to have been of extraordinary 
lengthy for the most part somewhere about forty years each. 
Even Newton expresses his opinion of the improbability of a 
succession of princes reigning for so long a period, and assigns 
to the reign of a king as a mean number, seventeen years. 
But the truest parallel is to be found in the case of the doges 
of Venice, who like the kings of Rome were elective princes; 
In a period of 500 years (a.d. 800 — a.d. 1300) Venice had 
forty doges, so that there were eight in each century. Xow if 
we closely examine the number of the Roman kings, we shall find 
a numerical artifice just as among the Orientals. I shall pre- 
mise the following considerations to illustrate what I mean. 

The Etruscans had, as the foundation of their chronology, 
two kinds o^ saecula, physical and astronomical; the latter con- 
tained 1 1 years, as the supposed mean number of the physical ; 
and by a double intercalation the calendar was restored so as 
to leave a wonderfully small difference. 110 of these years 
were nearly equal to 132 years, of ten months each, and this 
consequently formed an astronomical period. The physical 
saeculum was thus defined by the Etruscans : the first saeculum 
was determined by the life-time of the person who lived the 
longest, of all those that had been alive at the foundation of a 
state, the second was indicated by the longest life of the persons 
living at the conclusion of the Qvst' saeculum, and so on. Kow 
we find an ancient tradition in Plutarch and Dion Casslus 


(Dionysius lias at least an allusion to it) tliat Xuma was born 
on the day of the foundation of Rome, so that probably his 
death in the year a.u. 77, determined the first saeculum of 
Rome.3 If this was the case wc see the reason why thirty- 
eight years (the number of the nundines in a year of ten months) 
were assigned to Romulus and thirty-nine to Numa. In regard 
to the last live kings there existed historical traditions, but they 
were not sufficient for the whole period. It was certain that 
Rome had had far more than five kings, and as there were still 
wanting one as the founder of the Ramnes and another as that 
of the Titles, a number was chosen which had a sacred meaning, 
namely, the number of the planets, etc. The first half of 240 
years is the end of the 120th, that is exactly the middle of the 
reign of the fourth among the kings, manifestly an artificial 
invention; twenty-three years were assigned to him in order 
to make them begin with the year 110, some striking number 
being always desired for the beginning of a reign and 110 
being the secular number. The ancient year had ten months, 
and 132 of such years are equal to 110 of the later ones; it was 
therefore necessary to place the reign of Ancus between 110 
and 132. The period between 77 and 110, or thirty-two years, 
was naturally assigned to Tullus Ilostilius. Tarquinius Priscus 
reigned until a.u. 170, half a century being added to half the 
years of the kingly period, and his reign accordingly lasted 
thirty-eight years. The twenty-five years of the last king may 
be historical ; but it is possible also that a quarter of a century 
was assigned to him. The period from A.U. 170 to A.U. 21o 
was left for Servius Tullius. But now, supposing that the two 
reigns of Tarquinius Priscus and Servius Tullius did not last 
so long, all absurdity disappears, and the ancient unanimous 
account that Tarquinius Superbus was a son of Tarquinius 
Priscus is restored to its full riglit. AVe see then how the 
greatest nonsense arises from chronological restorations; the 
forgery is manifest. 

Now although the other sources of the earliest history, the 
ancient lays, were not falsified, they are nevertheless entirely 
insulficieut. We have a parallel to this in our own lay of the 
Nibelungen ; its authors have no intention to deceive ami do 
not pretend to give an annali^tic history; historical persons 

* T. Tiitius is said to have given him his tlaughtcr in niarriago, ami yet 
Tatius dies in the fourth yeai" after the fouudation of Kome. — N. 


occur in it such as Theodoric, Attila, the Burgundians, and 
yet no one portion of the whole poem belongs to history. In 
like manner, history cannot claim Romulus and Xuma, they 
belong to the sphere of the gods, Romulus as the son of jNIars, 
and Xuma as the husband of Egeria ; Romulus is only a per- 
sonification of Rome. Other poems of a similar kind contain 
more of historical substance, such as the Spanish Romances of 
the Cid; in this the fundamental features are indeed historical, 
but they form only a line, whereas the substance as given in 
the poem is a surface. It is the same with many portions of 
Roman history, and whoever entirely rejects the early history 
of Rome does not know what he is doing. Romulus and Xuma, 
then, form the first saeculum, because they do not belong at all 
to history; they form a saeculum by themselves, as it were a 
totally diiFerent period ; and whatever ancient traditions were 
found respecting the succeeding kings and their period (and 
many such traditions were current) were inserted in the chrono- 
logical outline. Any who may think this criticism dangerous, 
would cease to do so, Avere they better acquainted with events 
nearer our own time. It is well known that the middle-age 
romances about Charlemagne and his Paladins are based upon 
Latin chronicles ascribed to archbishop Turpinus; these we now 
look upon as romance and allow them to stand by the side of 
history; but who would believe that scarcely 150 years after 
Charlemagne, in the reign of Otho the Great, when not even 
the remotest idea of a crusade existed, the chronicle of Bene- 
dictus of Soracte gives a detailed account of an expedition of 
Charlemagne to Jerusalem, and without any suspicion of its 
not being true. Even before the Carlo vingian race was extinct, 
we find wholly fabulous features in the history of Charlemagne, 
such as his journeys across the Alps, etc., related in the chro- 
nicles with the greatest possible assurance. These we can now 
refute, as we have contemporaneous annals and the biography 
of Eginhard; the expedition to Jerusalem is disproved even 
without these by Oriental annals. It is the same in Ireland, 
for there too we find annals in which a series of kings is given, 
and among them Xiall the Great, a contemporary it seems of 
the emperor Theodosius; he conquered Britain, Gaul and Spain, 
crossed the Alps and threatened the emperor in Rome. The 
most positive evidence can be adduced against this entirely 


fabulous account, for the autlientic liistory of that period is 
generally known.* 

We might with the same facility prove that the early his- 
tory of Rome is not authentic, if we had earlier historical books 
to correct the legends. But where are we to find them? The 
Greeks did not come in contact with Rome till long afterwards, 
and although they possessed information about the Romans at 
an earlier period than is commonly supposed, they nevertheless 
gave themselves no concern about them, just because they did 
not come in contact with them. The case might be different 
in regard to the Greeks of the south of Italy and the Siceliots, 
but none of their writers have come down to us: neither Hero- 
dotus nor Thucydides could make mention of the Romans. 
But there still exists an isolated fragment of Etruscan history, 
which gives us an opportunity of seeing the manner in which 
the history of Rome was told among other nations. The em- 
peror Claudius, who was so unfortunate in his early youth and 
so ill used by his mother, and whose weak mind, although he 
was possessed of many amiable qualities, was entirely misguided 
by bad treatment, seems to have excited the sympathy of Livy, 
who instructed and encouraged him in historiography. He 
accordingly wrote several works in the Greek language, 
Kap)(^t]SoviaKd in eight and TvppriviKa in twenty books, the loss 
of which we have great reason to regret. Even Pliny does 
not notice the last named work. But in the sixteenth century 
there were found two tables, containing fragments of a speech 
of the emperor Claudius, in which he proposes to the senate, 
to grant the full franchise to the Lugduncnsian Gauls and to 
admit them into the senate as had long been the case in the 
provincia Romana. The inhabitants of Gaul were Roman citi- 
zens and had Roman names, but they had not the right to be 
admitted into the senate; and it was this right that the emperor 
Claudius conferred upon the Lugduncnsian Gaids. Of the 
several brass tables which contained the speech mentioned by 

■• The old Ii"ish tradition, as far as I can ascertain, diflcrs somewhat from 
the statement made in tlic text. It was not Niall tlie Great who advanced as 
far as the Aljis, l)ut his successor Dathy, w!io was struck dead at the foot of the 
Alps by a tiasli of lightning a.d. 427. Comp. Keating's General History of 
Ireland, translated by Derniod U'Conor. Lond. 1723, foL p. 319; M'Derniot's 
History of Ireland, Lonilon, 1820, 8vo. vol. i. p. 411. Tlic accounts of Roman 
■wTiters on Ireland, ai-e collected in O'Conor's Rerum Hibernicarum Scriptores, 
V. i., Prolcgom. p. 1.— Ed. 


Tacitus two still exist; they do not contain a continuous por- 
tion of tlie speech unless a considerable piece is wanting at the 
bottom of the first table. Previous to the French revolution, 
they were kept in the town hall of Lyons, but whether they 
are still there I cannot say.^ They give us an idea of Clau- 
dius's stupidity and we must acknowledge that the ancients 
did not wrong him in this respect. In this speech he says in 
detail what Tacitus has compressed into a few words. " It 
ought not to be objected," says the emperor, " that this is an 
innovation, since iiniovations have been made ever since the 
beginning of the state ; strangers have always been admitted, 
as for example the Sabines of T. Tatius ; strangers have even 
been made kings, to wit Numa, Tarquin the Etruscan, a de- 
scendant from Greece, and Servius Tullius, who according to 
our annals was a native of Corniculum , and according to those 
of Etruria an Etrurian of the name ofMastarna, and a follower 
of Cacles Vibenna. He migrated, settled on the Caelian hill, 
which was thus called after his leader, and there called himself 
Servius Tullius." This then is a direct proof of what the 
Koman annals were in those days. For nothing that is related 
of this Etruscan Mastarna can be applied to Servius Tullius, 
the son of a female slave. 

There is therefore no doubt that the earliest history of Eome 
arose ovit of lays. Perizonius mentions similar instances among 
other nations: even in the historical books of the Old Testa- 
ment such lays are to be found ; in reference to the Romans he 
quotes as a proof Cato's testimony, to which Cicero refers in 
two passages: "Would," says Cicero, "that those lays were 
extant, which Cato in his Origines states used many ages before 
his own time to be sung at repasts by the guests in praise of 
illustrious men." A third mention of them is found from Varro 
in Nonius Marcellus to the effect that jmeri honesti sang at re- 
pasts songs in praise of deceased great men, sometimes with 
and sometimes without the accompaniment of the flute. Every 
one must consider these testimonies to be valid. Among all 
nations with whose early national literature we are acquainted, 
we find either long historical poems of an epic character or 
short ones in praise of individual men. Now previously to 
making and proving the assertion, that fragments of both kinds 

* They are printed in Lipsius' edition of Tacitus and in Gruter's Corpus 
Jnscriptionum, Imt ai'c little read. — N. 


have come down to us in Eoman history, I must make some 
remarks upon the oldest metre. 

The Ancient Romans, before their adoption of Greek poetry, 
used the Saturnian verse, of which Horace speaks; 
Horridus ille dcfluxit numerus Satumius, 

and wliich several ancient grammarians have explained. Atilius 
Fortunatianus and others among them, being ignorant of its 
real nature, confined their remarks to a couple of lines that 
were extant, especially to the following: 

Maliun dabunt Metelli Naevio poiitae, 

in which according to the opinion of the time a hypercatalectic 
senarius appears. Terentianus Maurus, who belongs to the 
end of the third centiuy, speaks of it in treating of the Anacre- 
ontic verse, because the first part of the Saturnian resembles it. 
But the true Saturnian verse is quite different, as I intend 
shortly to show in a sej^arate treatise. It is capable of a variety 
of forms and is quite independent of Greek metres. The Latin 
expression for rhythm, which was not applied to Greek metres 
till a later time, is nimieri. The Greek metre is based upon 
music and time, but the Eomans actually counted the syllables 
and rarely if at all measured them ; a certain number of syl- 
lables was necessary to constitute rhythm. Our forefathers too 
had no idea of long or short syllables after the Greek fashion; 
in the old hymns of the Latin Church likewise short syllables 
are used as long and vice versa. Plautus and Terence in their 
iambic and trochaic verses in reality observe the rhythm only 
and not the time. The same is the case w^ith all Northern 
nations. The prevailing character of the Saturnian verse is, 
that it consists of a fixed number of feet of three syllables each. 
The number of feet is generally four, and they are either 
bacchics or cretics, alternating with spondees. Sometimes the 
erotics predominate and sometimes the bacchics; when the 
verses are kept pure the movement is very beautiful, but they 
are generally so much mixed that it is dilHcult to discern them. 
These verses, in use from the remotest times, are quite analo- 
gous to the Persian, Arabic, the ancient German, Northern 
and Anii'lo-Saxon verses, and in fact to all iu which alliteration 
prevails. The old German verse is divided into two halves, 
an alliteration occurs in the first half twice and in the second 


lialf once ; it lias four arses. The same fourfold rliytlim occura 
in the old Saxon harmony of the gospels, in Otfrid and others, 
but five or even six rhythms may occur; in the Persian we find 
generally four feet of three syllables, in the Arabic frequently 
the same, but often also feet of four syllables. The Spanish 
coplas de arte major which were common previous to the 
adoption of the Alexandrines, and which were introduced into 
Flanders, also are of exactly the same kind. It is probable 
that the same metre is found in the longer Provencal poems. 
This ancient Roman metre occurs throughout in Roman poetry 
down to the seventh century. I have collected a large number 
of examples of it and discovered a chapter of an ancient gram- 
marian with most beautiful fragments especially from Naevius. 
I shall publish this important treatise on the Saturnian verse, 
for the grammarian really understood its nature.^ In Plautus 
it is developed with great beauty. 

There were also smaller ancient poems in this metre. At 
the funerals of Romans naeniae were sung with the accom- 
paniment of the flute, and these were not melancholy and soft 
dirges, but must have had the same character as the laudationes; 
the dead had passed to their illustrious ancestors, their glory 
was made use of as a show and as an encouragement, and for 
this reason simple praise was bestowed upon them in these 
naeniae. The words of Horace, absint inani funere naeniae, etc., 
refer, if songs were sung at all at funerals, to the lamentations 
of later times; for the Romans originally were not tender- 
hearted : they made use even of a dead man for the good of the 
republic; from his grave he continued to call upon the living 
to follow in his footsteps. Naeniae and laudationes, therefore, 
were certainly quite plain and simple, according to the ancient 
style in which periods were not j^et known, and bore no 

^ The grammarian, whose fragment on the Satiirnian verse is here mentioned, 
is Charisiiis. Niebnhr took a copy of it from a Xcapolitan manuscript in 1823, 
and his copy has been entnisted to Prof. Lachmann of Berlin, who is preparing 
its publication. Prof. Schneidewin of Giittingen publislicd it in 1841 in a pro- 
gramme, "Flavii Sosipatri Cliarisii de versa Satm-iiio commentariolus ex codice 
Neapolitano nunc primiuu editus," from a copy taken by O. ^Mullcr, and severely 
criticised Niebuhr's expressions respecting the Saturnian verse; but a glance at 
the fragment, as it is there printed, shews, that Mliller's copy is very imperfect, 
and it would have been more becoming accurately to examine the copy taken 
by Niebuhr, before criticising him in a manner, which does not indeed injure 
the memory of Xiebuhr, but certainly does not place the modesty of Schneidewin 
in the most Aivourablc light. — Ed. 


resemblance to the X0704 errnd^Loi of Thvicydldes and tKe later 
Greeks. Two poems evidently of this kind are still extant on 
the tombs of the Scipios, which were discovered in 1 780 on 
the Appian road ; the upper compartment, which contained the 
sarcophagus of the yoimger Africanus and the statue ofEnnius, 
had disappeared, but the lower one was worked into the rock 
and was found filled with rubbish. The latter contained the 
sarcophagus of L. Scipio Barbatus, who was consul in the year 
A.U. 454. Persons had descended into this tomb from above 
lonjr before 7, and had taken out one of the slabs, which is now 
fixed in the wall of the palace Barberini, but it was forgotten 
again/' These magnificent sarcophagi bear inscriptions in 
verse, which are written like prose it is true, but the verses 
are divided by lines ; on the sarcophagus of the son the verses 
are even marked, and that they are verses may be seen from 
the unequal length of the lines, for otherwise the Eomans always 
wrote their lines to the end of the slab. These are quite plain 
and simple verses but still there is rhythm in them — 

Corneliu' Luciu' Scipio Barbatus, 

Gnaivo prognatu', fortis vir sapiens que — 

Consul, censor, aedilis, qui fuit apud vos, etc. 

— These are certainly the naeniae which were sung at the time 
and were afterwards inscribed on the tomb. The ancient songs 
at repasts were for the most part just as simple. 

Now these naeniae^ which together with the laudationes were 
kept in the atrium, are sources of the earliest history. But 
besides these there also existed longer epic poems among the 
Komans no less than among other nations such as the Servians ; 
the songs of the modern Greeks are of a purely lyrical character, 
but those of the Servians are a combination of epic and lyric. 
I think I have discovered in Livy a fragment of such an heroic 
epic, on the fight of the Iloratii and Curiatii. Xow we cannot 
indeed suppose that Livy saw these ancient epics and -wrote 
his history from them, but he wrote in part directly and in part 
indirectly through the medium of Varro, from the books of the 
pontiffs and augurs, w^hich contained a great many fragments 
of such ancient epics, some of which may have been as old, 

^ In the year 1616. 

« The bodies of the Cornelii down to the time of Sulla were not burned 
according to the Pelasgian and Greek fashion, but were buried in coffins.— N. 


even as the time of the taking of Rome by the Gauls. In the 
passage of Livy in which he relates the trial of Horatius, which 
he took from those books, he speaks of a lex horrendi carminis; 
the formulae of that time were called carmina and were in the 
ancient metre. That Livy drew his materials from those books 
either directly or indirectly becomes the more certain from 
Cicero's statement, that the formula of the ^;rot;oc«/?o ad 'pojmlwn 
was contained in the lihri augurales. The formula is — Duum- 
viri perduellionem judicent, etc., in which the ancient metre is 
still discernible. 

I have elsewhere observed that Cicero's statement : lauda- 
tionibus historia nostra facta est mendosior, is also acknowledged 
by Livy: as every thing good may easily acquire a tendency 
to evil, so also could the beauty of Eoman family pride dege- 
nerate into falsehood, and there is no reason for disbelieving 
the assertion. 

After the first scanty records of the early times had for the 
most part been destroyed in the Gallic conflagration, they were 
restored according to certain schemes from the songs of the 
vates; the poems became altered as they passed from mouth 
to mouth, and they, combined with the laudationes, form the 
groundwork of our history— the material which Fabius found 
when he began to write. 

If we look at the tenth book of Livy, we find in it a 
disproportionate minuteness in his account of the campaigns 
of Fabius Maximus liulllanus, and this minuteness arises from 
family records; we may in fact point out not a few statements, 
which cannot have had any other source but family vanity, 
which went so far as to forge consulships and triumphs, as 
Livy himself says. 

Otlier forgeries again arose from national vanity, and these 
occur everywhere in those parts of the history which relate to 
any great calamity suliered by the Romans, especially the great 
calamities of the early times, such as the war with Porsenna, 
the sacking of Rome by the Gauls, and the defeat of Caudium, 
the whole narratives of which are falsified. Others arose from 
party spirit, which in primitive periods led to perpetual strife; 
one party raised false accusations against the other, and these 
were introduced into history; at other times attempts Avere 
made to palliate and conceal moral and political crimes. The? 
people are described as being the cause of the worst misfortunes 


tLougK they were innocent and their opponents were the guilty 
party; it was not the people but the curiae that condemned 
Manlius to death, and it was the curiae that pronounced the 
inglorious decision between the Ardeatans and Aricinians; nay 
we may be convinced that it was the curiae too who compelled 
Camillus^ to go into exile. 

Such falsifications accumulate, become interwoven with 
one another, and in the end produce a strange confusion. We 
may collect the rich materials though they are widely scattered, 
because party spirit prevented their being united, and by the 
process of criticism we may discover the constitution and 
character of the Roman nation, and in general outlines give 
their history down to the time at which we have the contem- 
porary records of the Greeks, that is to the war with Pyrrhus 
and the first Pimic war. j\Iuch Avill indeed remain obscure in 
our investigations, but we can accurately distinguish where 
this must be so and where not. 

Roman History goes back to Latium and through Latium 
to Troy. Since the question was raised by Dion Chrysostomus 
whether Troy ever really existed, an immense deal has been 
written upon it, and also on the question whether Aeneas ever 
came to Italy. The treatise by Theodore Ryckius^" upon this 
subject is very well known; he regards the arrival of Aeneas 
as an historical fact in opposition to Bochart, who was one of 
the last ingenious philologers of France", and whose Intellect 
was at all events superior to that of Ryckius. Bochart's 
hypothesis concerning the influence of the Phoenicians is cer- 
tainly carried too far. Now, however, the question would be 
put in a totally difTerent manner, we should ask, Plas the legend 
of the arrival of the Trojans on this coast any historical ground? 
Further, Did the legend originate with the Greeks and come 
over to Italy, or is it of native Italian growth, that is to say, 
is it one which we at least cannot trace to any Greek sources ? 
If the latter be the case, there must be some truth at the bottom 
of it, and the less we take these ancient traditions literally, the 
more probability we find in them. 

There existed unquestionably in the earliest times of Greece, 

' Li\7, iii. 71, 72. 

"> Thcod. Ryckii l)if;s. de Primift lUiliae Colonis ct Acnea in Luc. Ilolstcnii 
Notae et Caxtigationes in Stepli. Hi/zantiiim. Lugd. Bat. 1684, fol. 
" Siilmasius was fax less clear-headed than he.— N. 


two nations wlio were very nearly akin to eacli other and yet 
were so different that the one did not even understand the 
language of the other, as Herodotus distinctly says : the lan- 
guage of the one when compared with that of the other was 
regarded as barbarous, and yet from another point of view they 
may be looked upon as very kindred languages. Several 
living languages, even now, stand in a similar relation to one 
another^ such as the Polish and Bohemian, the Italian and 
Spanish, and if we do not look at the relationship quite so closely, 
the Polish and Lithuanian. The last two languages differ 
from each other immensely, but yet have a characteristic re- 
semblance; the grammar of both is based upon the same prin- 
ciples: they have the same peculiarities, their numerals are 
almost the same and a great number of words are common to 
both. These languages therefore are sister languages and yet a 
Pole does not understand a Lithuanian. Now this is the man- 
ner in which we solve the question so often raised respecting 
the difference or identity of the Greeks and Pelasgians. When 
Herodotus tells us that they were different, we must indeed 
believe him, but on the other hand he joins the Hellenes and 
Pelasgians together, consequently there can have been no radi- 
cal difference between the two nations. 

In the earliest times, when the history of Greece is yet 
wrapt up for us in impenetrable mystery, the greater part of 
Italy, perhaps the whole of the eastern coast of the Adriatic, 
Epirus, Macedonia^*^, the southern coast of Thrace with the 
peninsulas of Macedonia, the islands of the Aegean as well as 
the coasts of Asia Minor as far as the Bosporus were inhabited 
by Pelasgians." The Trojans also must be regarded as Pelas- 
gians; that they were not barbarians is confirmed by the 
unanimous opinion of all the Greeks and may be seen from 
Homer; they inhabit a Pelasgian country but their names are 
Greek. They are sometimes spoken of as more closely con- 
nected with the Arcadians,, who were another essentially Pelas- 
gic race, sometimes with the Epirots and sometimes with the 
Thessalians; Aeneas in one tradition migrates to Arcadia 
and there dies, and in another he goes to Epirus where Helle- 
nus is settled. Thus, in Pindar's poem on Cyrene, we find 

" The original inh;ibitants of jNIaccdonia Avcrc neither lilyrians nor Tliracians, 
but Pelasgians. Comp. C. O. Miilicr's Treatise on Macedonia, appended to 
"Vol. I. of the Hist, and Ant. of tlic Doric Race, p. 467, etc. 

'^ Even Aeschylus peoples all Greece with Pelasgians. — N. 


Arlstaeus, a Pelasgian hero from Arcadia, along with the 
Antenoridae. The connection between the Pelasgians and 
Trojans goes very far back, for Samothrace especially is the 
metropolis of Ilium; Dardanus comes from Arcadia, but passes 
through Samothrace, and, being married to Chryse, he proceeds 
thence to Troas. The Samothracians, according to one gram- 
marian, were a Roman people, that is, they were recognised as 
the brothers of the Romans, namely of the Troico-Tyrrhenian 
Pelasgians. This connection has no other foundation than the 
kindred nature of the Tyrrhenians, Trojans, and Samothra- 
cians. Some accounts state that Dardanus went from Tyr- 
rhenia to Troas, others that the Trojans went to Tyrrhenia. 
The temple and mysteries of Samothrace formed a point of 
union for many men from all countries ^^: for a great portion 
of the world at that time, the temple of Samothrace was like 
the Caaba of Mecca, the tomb of the prophet at Medina, or 
the holy sepulchre at Jerusalem. Samothrace and Dodona 
were to the Pelasgian nations what perhaps Delj)hi and Delos 
were to the Hellenic world. Tiie distance of a great number 
of kindred tribes from those central points, was in this instance 
of no greater consequence than in the case of the I\Iahomme- 
dans, who are not prevented by distance from going as pilgrims 
to the sacred spot. 

This race of the Pelaso-ians, which we can trace as far as 
Liguria, and which also inhabited at least the coasts of Corsica 
and Sardinia, disappears in the historical times as a body of 
nations: it consisted originally of a number of tribes with 
different names, of which afterwards we find only remnants 
and isolated tribes. A very extensive name for that part of 
the race which inhabited Epirus and the southern part of 
modern Italy, at least as far as Latium and the coast of the 
Adriatic, was Siculi, also Vituli, A'itelli, Yitali Itali; from 
these Italy derives its name.^^ Notwithstanding the wide 
extent of this Siculian or Italian name, it seems that in the 
earliest times Italy did not, as now, denote the country as flir 
as the Alps; it is indeed possible that the changes which took 

'■' Wc may certainly look upon this as an established fiact, althoii^h the in- 
vestigations conecvning the mysteries themselves will never yield any positive 
results. — N. 

'^ As K (or C) and T arc identical, and oidy dialectically different, so tlie S is 
changed into the digamma or Y, which, again, is often lost, especially at the 
beginning of words.— N. 


place in consequence of tlie migration of the northern tribes se- 
parated the maritime countries of Etruria from Italy and confined 
the nameof Italy to the country south of the Tiber or even south of 
Latium. This, however, is only a conjecture; but it is certain 
that at one time Italy was bounded in the north by a line from the 
Garganus in the east to Terracina in the west, and that the name, 
after having been more limited, was again, after the time of Alex- 
ander the Great and previously to the extension of the dominion 
of Rome, used in its former and wider extent. It seems to be this 
earlier Italy that F liny means, when he says it is querno folio 
similis}^ This statement he undoubtedly took from Timaeus,with 
whom also originated the comparison of Sardinia to a sandal or a 
foot-mark. It quite escaped Pliny's attention that Italy in his 
time could not be described in any such way ; and this is a 
very characteristic instance of the hasty and thoughtless 
manner in which he wrote. 

In the south of Italy the earliest inhabitants were also 
called Oenotri and Peucetii, in the north undoubtedly Libur- 
nians and on the coast of Latium TyiThenians. 

Whether the settlements on the coast north of the Tiber 
were remnants of a people who had been driven back, or 
whether they were only colonies, it is no longer possible for us 
to decide. But there appear in central Italy besides these 
tribes, which were analogous to the Greeks, nations of a different 
kind which overwhelmed the former. These migrations seem 
to have been similar to those met with in modern history, 
where one nation has pushed forward another. The people 
who threw themselves at the same time upon the Siculi in 
Latium and upon the Itali in the south of Italy, and, having 
partly expelled and partly subdued them, became assimilated to 
them, are the Opici, a transition people, who in reality existed as 
Opici in a few places only, but, being again amalgamated with 
other subdued people, they produced new forms. They appear 
under various names, which, however, have the same radical 
syllable. Thus we find them under the name of Apuli , the termina- 
tions -icus and -ulus being equivalent : hence the Italian population 

'® This is a remarkable example of the manner in which Pliny wrote; he 
sometimes speak in his own name and sometimes gives extracts, but nnfortu- 
nately his historical extracts arc made with as little thonght as those relating to 
natural history, which arc full of misapprehensions of Aristotle and Theophras- 

tus.— N, 



ceases in Apulia, extending apparently as far as Messapia, 
where a portion of the Itali maintained themselves in an iso- 
lated position. They further existed in the countries after- 
wards called Samnium, Campania, and, under the name of 
Volscians and Aequians, on the borders of Latium. 

The Opicans again were pressed forward by the Sabines 
(Sabellians) who called themselves autochthons, and who 
traced their origin to the highest mountains of the Abruzzo, 
near Majella and Gran Sasso d^Italia. Cato somewhat 
strangely supposes these to have come from the small dis- 
trict of Amiternum. Xow whether the Sabellians and Opicans 
diftered from each other, as, for example, the Gauls and Li- 
gurians did, or even in a less degree, as the Gauls and Cymri; 
or whether they belonged to the same stock and were sepa- 
rated from each other only politically, are questions which we 
cannot solve. The ancients did not know this, nor did they 
pay miich attention to it. If we obstinately determine to see 
where no historical light is to be obtained, the intellectual eye 
is injured as is the physical, when it violently exerts itself in 
the dark. Varro indeed distinguishes between the Sabine and 
Oscan languages, but he knew so little of the ancient languages, 
in the sense in which \\\ v. Humboldt knows them, that little 
reliance can be placed on his statements respecting the affinity 
of languages. According to general analogy, I believe that 
there was a migration of nations in different directions, by the 
first impulse of which the Sabines may have been driven from 
their northern habitations; but this is a mere conjecture. 

The Umbrians may possibly have belonged to the same stock 
as the Opicans. I should not like to attribute too much im- 
portance to the resemblance of their names, for nations that 
are nearest akin to one another often have very diifcrent names, 
and widely different nations frequently have similar ones. 
Thus the Getae and the Goths were for a long time errone- 
ously looked upon as the same people ; and fifty years ago it 
was the general opinion in Ireland and Scotland that the Fir- 
Bolgs^^ spoken of in the poems of Ossian were the ancient 
Belgians. But this is not correct ; they were, as a very well 
informed Englishman wrote to me, a Danish colony. "We must 

" The Fir-Bolgs belong to the bardic history of Irelaiul, which describes them 
as the tliird immigration into Irehmd ; the Scots found them in Ireland governed 
by kings ; to them is ascribed the building of the Cyclopian walls in Ireland. — Ei\ 
VOL. I. C 


be greatly on our guard against the miserable desire to construe 
the history of nations from their names, a desire which has 
given rise to so many hypotheses and fancies. Much may be 
learnt from the study of names indeed, but' what in some cases 
is correct ceases to be true in others, and becomes a source of 
error and fanciful theories which we must shun as vermin and 
serpents. If I had not other evidence than the mere names, I 
should hesitate to declare the Opicans and Umbrians identical. 
But Philistus called tlie people who conquered the Siculians in 
Latium Ombricans, and moreover the affinity of their languages 
may be distinctly perceived from the remnants which have 
come down to us. 

These changes of nations, in which the earliest inhabitants 
were driven out by one tribe and this again by another, are 
the causes which render the history of the early Italian nations 
so indescribably obscure and difficult for us, that, even where 
we ourselves have a clear view, the misconceptions in our 
authorities still maintain their ground, and ever and anon cause 
fresh discussions. A solution of these difficulties, free from all 
objections, is utterly impossible. He who is engaged in such 
investigations must often be satisfied with evidence, which has 
the appearance of truth, but he ought to be able to shew how 
the misconceptions arose. 


At a period which we cannot chronologically define, there 
existed a popvdation of Siculians in the country afterwards 
called Latium, which may however have borne this name from 
the earliest times. The remembrance of this population was 
preserved at Tibur, part of which town was, according to Cato, 
called Siculio.^ Elsewhere also in ancient authors, we find an 
immense number of statements which place the existence of 
this people beyond all doubt. It is found under the same name 

" In the printed collections of the fi-agments of Cato, I do not find this state- 
ment ; whence I suppose that Cato is here confounded with Dionysiuswho(i.l 6) 
has the statement in question. — Ed. 


in southern Italy, and also in tlie island wliich to tliis day is 
named after them. According to one tradition, Sicelus went 
from Latium to the Oenotrians; according to another, the 
Siculians under different names were driven from their ancient 
habitations by the Opicans or Umbrians, and migrated to the 
island of Sicily, This migration only shews the combinations 
of those who wish to prove the contemporaneous existence of 
the same people in Latium and in Sicily. The migration is pos- 
sible indeed, but it is also possible that it took place in quite a 
different direction. It is certain that the Siculians existed in 
the south of Italy in Homer's time, of which we find evidence 
in a passage from ]\Inaseas, a pupil of Aristarchiis, a learned 
grammarian and historian quoted by the Scholiast on the 
Odyssey. He says also, that Echetus was prince of the Siceli 
in Epirus, so that he recognises this name even in those parts ; 
we see from his explanation, that when the poet of the Odyssey 
speaks of the Siceli, he does not mean the inhabitants of Sicily, 
an island scarcely known to him, but the inhabitants of the 
south of Italy or the Pelasgians of Epirus. 

The Siculi are the same as those whom Cato calls Abori- 
gines. This name is explained by yevdpxai, that is ancestors ; 
or by Aberrigines, that is, icandering people ; but it more pro- 
bably signifies the people that have been from the beginning 
(ab origine). The nominative singular according to the Latin 
idiom must have been Aboriginus. There was a tradition that 
Latium was originally inhabited by autochthons, but Cato 
and C. Sempronius- said, that the Aborigines had come from 
Achaia, that is from the Peloponnesus, the whole of which 
was then called Achaia by the Romans. Others apply the epi- 
thet Argive to the particular places which were otherwise 
called Siculian, and Cato had done so even in the case of Tibur. 
Argos and Larissa are Pelasgian names occurring wherever 
Pelasgians are found, Argos probably signifying a toii-n, and 
Larissa a citadel or arx. So long as the Peloponnesus was Pe- 
lasgian it was called Argos, just as Thessaly and, in this sense, 
the Arglves are Pelasgians ; the 'Apyelot IleXaa-jol in ancient 
tragedy arc always mentioned together, the one being probably 
the wider and the other the more limited name. 

Hesiod says of Latinus, Traai Tvp'pi^volaiv ciyaKXeiTolatv 

' Probably C. Sempronius Tuditanus, the same whom Dionysius, i. 11, calls 
KoyidTarov rwv 'Pa'/io(o>v avyypa<pi(iiv. — Ed. 



dvdaaei. All we know about the Latins is the fact that they 
possessed a number of towns from Tibur to the river Tiber: 
how far they extended in the earliest times towards the Liris 
is uncertain. Cato, quoted by Priscian, states that the plain of 
the Volscians formerly belonged to the Aborigines; and it is 
certain that all the towns along the coast, such as Antium, 
Circeii and others, were at an early period Tyrrhenian. At 
that time, accordingly, the name Latium was of very wide ex- 
tent, and even immediately after the time of the Roman kings 
it extended as far as Campania, but was afterwards restricted 
by the great migrations which took place after the expulsion 
of the kings. Hesiod of course refers to an earlier period. In 
the treaty of Rome with Carthage the names Latium and 
Latins extend along the coast beyond Terracina, and probably 
as far as Cumae. 

The Pelasgian inhabitants of the whole of the western coast 
of Italy were called by the Greeks Tyrrhenians, and by the 
Latins Turini, Tusci, that is, Tusici from Tusus or Turus; for* 
is used in the early language instead of r, as in Fusius for 

We must keep in mind that the Pelasgians and Aborigines 
were one and the same people. If we examine the tradition^, 
of nations we frequently find that the same events are related 
in various and entirely opposite ways. The story of a Jew 
taking merciless vengeance on a Christian, such as we read of 
in " The Merchant of Yenice", is foimd completely reversed 
in a Roman tale written shortly before Shakespeare's time; in 
this the Christian is represented as wishing to cut a piece of 
flesh out of the Jew's body. The migrations of the Goths 
proceed, according to some, from Scandinavia to the south, and 
according to others from the south to Scandinavia. Wittekind 
states that the Saxons came from Britain to Germany, while 
the common tradition describes them as having been invited 
fi'om Germany to Britain. The Pelasgians about Mount 
Hymettus near Athens are said to have migrated from Tyr- 
rhenia to Athens, and thence to Lemnos, while in another 
tradition the Tyrrhenians proceed from the INIaeonian coast to 
Italy. In like manner, Cyrene, according to one tradition, 
received a colony from Thera ; but according to another, Tliera 
arose out of a clod of earth from Libya. In the earlier tradi- 
tions, the Planetae are at the entrance of the Euxine, and the 


ship Argo on its voyage to Colchis sailed between them : in 
the later traditions, they appear in the Western sea and are an 
obstacle to the Argo on her return. The same contradiction 
appears in the case of the Aborigines. Dionysius in defiance 
of etymology applies this name to the people, who coming from 
the interior overpowered the ancient inhabitants. Varro did 
just the same: he is even worse than Pliny; he knows that the 
Latins are a combination of two nations, but he confounds 
every thing, representing the Aborigines as the conquering 
and the Siculians as the conquered people.^ Following the 
example of Hellanicus he proceeds to trace the Aborigines to 
Thessaly, but then makes them migrate from the Upper Anio 
as far as the Upper Abruzzo, whither they are pushed by the 
Sabines. This tradition has a local and probable character, 
for in that district there existed a number of small townships ; 
large towns on the other hand, such as we find in Etruria, are 
always a proof of immigration^ the immigrating people usually 
settling together in considerable numbers. Dionysius must be 
excused for his error, since he trusted to the authority of Varro, 
who alone is responsible for the blunder of confounding the 
conquering with the conquered people. 

One of the conquering tribes probably bore the name of 
Casci. Whether this was one of the names borne by the Tyr- 
rhenians, Latins or Siculians, or whether the Casci were foreign 
immigrants, cannot be determined with certainty, though the 
latter is more probable. The name Casci has been preserved 
by Servius from Saufeius, a grammarian who seems to belong 
to the first century of the Christian era. They also occur under 
the name of Sacrani, from which Varro and Dionysius infer that 
they were a lepa V€6r7)<;. A tribe of the people who, under the 
names of Opicans, Oscans and Umbrians, inhabited the interior 
of Italy, or, more probably, had been pushed forward from the 
north and was pressed between the ancient Pelasgian places, 
settled in the Apennines about lake Fucinus (now Celano) 
towards Reate. 

Their capital was called Llsta, and they extended to the 

* Varro had read immeiisi.'ly, Init he oiiglit not to be called a learned man, 
on account of his confusion. A\nien I, as a young man, began these investiga- 
tions, I could not sec my way clearly in these matters though in the nu\iu jioints 
I saw correctly, I trusted too much to Varro's authority, and owing to his 
confusion of names I did not gain a clear insight till when I prepai'cd a new 
edition of my work. — N. 


boundaries of the Sicvilians who dwelt above Tibur towards the 
inland districts. There was a tradition that in their war with 
the Sabines, who had already taken Reate from them, and 
continued to push them onward, they had vowed a ver sacrum. 
This custom, observed by the Italian nations in times of mis- 
fortune, was preserved among the Romans also : a vow was made 
to dedicate to the gods all the cattle and in general every thing 
which the next spring might produce, and to send out as 
colonists the male children who were born in that season ; the 
vegetable produce was either offered as a sacrifice or its value 
in money. Having made this vow, the Sacrani marched to- 
wards Latium and subdued the Siculians. In Latium they 
settled among the ancient inhabitants, and became united with 
them into one people bearing the name Prisci Latijii, for the 
Casci must also have been called Prisci^ Prisci Latini is the 
same as Prisci et Latini, for the Latin language always expresses 
two ideas which are inseparably connected by the simple juxta- 
position of the two words, mortar not being used by the ancient 
Romans in their language any more than in their architectural 
works. This has been clearly demonstrated by Brissonius who 
has also established the foxmu\-d jjopulus Romanus Quirites; bvit 
he goes too far in asserting that the Romans never said populus 
Romanus Quiritium, a position which has been justly contro- 
verted by J. Fr. Gronovius. In like manner we must explain 
patres conscripti as qui patres quique conscrijiti sunt, and also the 
legal formulae, locati conducti, emti venditi, and others. Priscus 
and Cascus afterwards signified venj ancient, old fashioned; 
whence the phrases, casce loqui, vocahula casca. These 
conquerors spoke Oscan, and from the combination of their 
languase with that of the Pelasgo-Sicvdians there arose that 
curious mixture which we call Latin, of which the grammar, 
and still more the etymology, contains so important a Greek 
element, which C. 0. Mliller has at my suggestion so admirably 
investigated in the first volume of his Etruscans. The primitive 
Oscan language is still preserved in a few ancient monuments; 
a few inscriptions in it were found at Pompeii and Hercu- 

* It would be absurd to take Prisci Latini in Li^y to mean ancient Latins; 
he took tlie formula of the declaration of war by the Fetiales, in wliich the 
expression first occur"!, from the ritual books; it refers to the time of Ancus 
Martins ; and before the time of Tarquinius Superbus there were no Latin colonies 
at all as distinguished from the rest of the Latins. — N, 


lanum ; the table of Bantia (Oppido) is perfectly intelligible. 
Of the two elements of the Latin language, the Greek and the 
not- Greek, the latter answers to the Oscan language. All 
words relating to agriculture, domestic animals, produce of the 
field, and the like, are Greek or akin to Greek. We see then 
a conquered agricultural people, and a conquering one comincr 
from the mountains, which did not pursue agriculture. 

Henceforth we lose all traces of the original tradition which 
is supplanted by the story of the Trojan immigration. I shall 
not here enter into any detail, but refer you to the minute 
investigations contained in my history of Eome; the result of 
which is that this last-named story has no authenticity what- 
ever, but is only a later embellishment to express the relation 
existing between the Trojans as Pelasgiaus, and the nations of 
Italy which belonged to the same stock. The tradition of a 
Trojan colony occurs in many parts of Italy, and the fact of 
its having become more firmly established in regard to Latium 
is purely accidental; it was kept up and nourished by the 
diffusion of Greek poems which was far more extensive than 
Ave commonly imagine. 

The story of the Trojan settlement is comparatively ancient 
among the Romans; even Xaevius, in his poem on the Pimic 
war, gave a very minute account of it ; the Ilians established 
their claims among the Eomans during the wars against Se- 
leucus Callinicus, We could not take as our guide a person 
who would treat seriously the accounts of the foundation of 
Rome by Aeneas; some particular points in them are of a 
really national character, but the period of time between the 
events and their recorders is too great. Naevius wrote about 
950 or 980 years after the time commonly assigned to the 
destruction of Troy. It is little known how much Vii'gil altered 
the ancient tradition of the settlement of Aeneas in Latium — 
as a poet he had a perfect right to do so — for its ancient form 
was rough and harsh, as Latinus was said to have fiillen in the 
war against Aeneas, and Lavinia who was first betrothed to 
Aeneas and afterwards refused him, became a prisoner of war. 
The earliest tradition, moreover, represented the settlement as 
very small, for, according to Xaevius, Aeneas arrived with only 
one ship, and the territory assigned to him consisted, as Cato 
stated, of no more than 700 jur/era. Supposing this to be true, 
how is it possible that a recollection of it should have been pre- 
served for upwards of 900 years? 


The original tradition is, that Aeneas at first for three years 
dwelt in a small town of the name of Troy; he is then said to 
have gone further inland and to have founded Lavinium ; thirty 
years after this, Alba was founded, and 300 years after Alba 
the foundation of Kome was laid. This regular progression of 
numbers shews that the field is not historical, and there seems 
to be no doubt that the duration of Rome was fixed at 3000 
years. There are in these traditions two different numerical 
systems, the Etruscan, with a saeculum of 110 years, and the 
Greek or Tyrrhenian in which the saeculum consisted of thirty 
years. This number thirty was at all times of great importance, 
because the period of the revolution of Saturn was then, as 
Servius remarks, believed to be completed in thirty years. 
Thirty ordinary years formed with the Greeks one Saturnian, 
and 100 Saturnian years constituted one great year. With this 
are connected the progressive numbers fi:om the foundation of 
Lavinium to the building of Rome. The earliest history of 
Alba is worth nothing, as has been shewn by the acute Dod- 
well^ ; who elsewhere too often spoiled by his subtleties that 
which he had well begun. The chronology of the Alban kings, 
for example, in Dionyslus is nothing but folly and falsehood, 
and their names are huddled together in every possible manner. 
This forgery, as we learn from Servius, was made at a late 
period by a freedman of Sulla, L. Cornelius Alexander of Mi- 
letus, who quickly became popular at a time when people 
delighted in having the history of a period of which nothing 
could be known. 

Alba on the Alban lake is, in my opinion, the capital of the 
ruling conquerors; it is not owing to mere chance that it bears 
the same name as the town on lake Fucinus whence the Sacrani 
had come. When they were obliged to give up their country 
to the Sabincs, they founded a new Alba on a lake, just as the 
Carthaginians built a new Carthage, the IMilesians a new M'l- 
letus on the Black Sea, and as the English have so often done 
in the new world. This Alba Longa then was the seat of the 
Casci or Sacrani, and the earlier Latin towns within its terri- 
tory probably experienced a twofold fiite; some may have 
received a part of their population from the immigrants, and 
others may have been reduced to a state of dependence without 
receiving colonists. We have a tradition that these Latin 
towns were thirty In number and that all were colonies of 

* De Cyclis, diss. x. 


Alba, but this is opposed to another statement which declares 
all of them to have been originally Argive towns. Both may 
perhaps be maintained, if we suppose that an a7roBaa/x6<; of the 
ruling people settled in each of the towns. This tradition as 
it stands is founded upon a misunderstanding : Alba had thirty 
dcmi, which as perioeci belonged to it, and they are the populi 
Albenses which I have discovered in Pliny. By this discovery 
their relation has become clear to me, and I have no doubt 
that the relation in which Alba stood to these Albensian towns 
was the same as that in which the populus of Rome stood to 
the plebs, and afterwards Rome to Latium. Previously to its 
destruction. Alba had no doubt the sovereignty of Latium, as 
Rome had afterwards. Alba therefore was surrounded by 
thirty populi Albenses, part of which were probably Alban 
colonies, and all of which constituted the state of Alba; and 
besides them there was a number of towns of the Prisci Latini, 
which were dependent upon Alba, whatever their condition 
may have been in the earliest times. 


I BELIEVE that few persons, when Alba is mentioned, can get 
rid of the idea, to which I too adhered for a long time, that 
the history of Alba is lost to such an extent, that we can speak 
of it only in reference to the Trojan time and the preceding 
period, as if all the statements made concerning it by the 
Romans were based upon fancy and error; and that accord- 
ingly it must be effaced from the pages of history altogether. 
It is true that what we read concerning the fovuidation of Alba 
by Ascanius, and the wonderful signs accompanying it, as well 
as the whole series of the Alban kings with the years of their 
reigns, the story of Numitor and Amulius and the story of the 
destruction of the city, do not belong to history ; but the his- 
torical existence of Alba is not at all doubtful on that account, 
nor have the ancients ever doubted it. The Sacra Albana and 
the Albani tumuli atque luci, which existed as late as the time 
of Cicero, are proofs of its early existence; ruins indeed no 


longer exist, but the situation of the city in the valley of 
Grotta Ferrata may still be recognised. Between the lake and 
the long chain of hills near the monastery of Palazzuolo one 
still sees the rock cut steep down towards the lake, evidently 
the work of man, which rendered it impossible to attack the 
city on that side; the summit on the other side formed the arx. 
That the Albans were in possession of the sovereignty of Latium 
is a tradition which we may believe to be founded on good 
authority, as it is traced to Cincius.^ Afterwards the Latins 
became the masters of the district and temple of Jupiter. 
Further, the statement that Alba shared the flesh of the victim 
on the Alban mount with the thirty towns, and that after the 
fall of Alba the Latins chose their own magistrates, are glimpses 
of real history. The ancient tunnel made for discharging the 
water of the Alban lake still exists, and through its vault a 
canal was made called Fossa Cluilia : this vault which is still 
visible is a work of earlier construction than any Roman one. 
But all that can be said of Alba and the Latins at that time is, 
that Alba was the capital, exercising the sovereignty over 
Latium; that its temple of Jupiter was the rallying point of 
the people who were governed by it; and that the gens Silvia 
was the ruling clan. 

It cannot be doubted that the number of Latin towns was 
actually thirty, just that of the Albensian demi; this number 
afterwards occurs again in the later thirty Latin towns and in 
the thirty Roman tribes, and it is moreover indicated by the 
story of the foundation of Lavinium by thirty families, in 
which we may recognise the union of the two tribes.^ The 
statement that Lavinium was a Trojan colony and was after- 
wards abandoned but restored by Alba, and further that the 
sanctuary could not be transferred from it to Alba, is only an 
accommodation to the Trojan and native tradition, however 
much it may bear the appearance of antiquity. For Lavinium 
is nothing else than a general name for Latium just as Panionium 
is for Ionia, Latiniis, Lavinus^ and Lavicus being one and the 
same name, as is recognised even by Servius. Lavinium was 
the central point of the Prisci Latini, and there is no doubt 
that in the early period before Alba rnled over Lavinium, 
worship was offered mutually at Alba and at Lavinium, as was 

' Albanos rerum potitos usque ad Tullum. Festus, s. v. prator. 
» Horn. Hint. i. p. 201, fol. 


afterwards the case at Kome in the temple of Diana on the 
Aventine, and at the festivals of the Romans and Latins on 
the Alban mount. 

The personages of the Trojan legend therefore present them- 
selves to us in the following light. Turnus is nothing else 
but Turinus, in Dionysius Tvpp7]v6<;; Lavinia, the fair maiden, 
is the name of the Latin people, which may perhaps be so dis- 
tinguished that the inhabitants of the coast were called Tyr- 
rhenians, and those further inland Latins. Since, after the 
battle of lake Regillus, the Latins are mentioned in the treaty 
with Rome as forming thirty towns, there can be no doubt 
that the towns, over which Alba had the supremacy in the 
earliest times, were likewise thirty in number ; but the confed- 
eracy did not at all times contain the same towns, as some may 
afterwards have perished and others may have been added. In 
such political developments, there is at work an instinctive 
tendency to fill up that which has become vacant ; and this 
instinct acts as long as people proceed unconsciously according 
to the ancient forms and not in accordance with actual 
wants. Such also was the case in the twelve Achaean towns 
and in the seven Frisian maritime communities; for as 
soon as one disappeared, another dividing itself into two, 
supplied its place. Wherever there is a fixed number, 
it is kept up, even when one part dies away, and it ever con- 
tinues to be renewed. We may add that the state of the 
Latins lost in the West, but gained in the East. We must 
therefore, I repeat it, conceive on the one hand Alba with 
its thirty demi, and on the other the thirty Latin towns, the 
latter at first forming a state allied with Alba, and at a later 
time under its supremacy. 

According to an important statement of Cato preserved in 
Dionysius, the ancient towns of the Aborigines were small 
places scattered over the mountains. One town of this kind 
was situated on the Palatine hill, and bore the name of Roma, 
which is most certainly Greek. Not far from it there occur 
several other places with Greek names, such as Pyrgi and 
Alsium ; for the people inhabiting those districts were closely 
akin to the Greeks ; and it is by no means an erroneous con- 
jecture, that Terraclna was formerly called Tpa-)(€ivi'], or the 
" rough place on a rock;" Formiae must be connected with 
opfMo<i, "a road-stead" or "place for casting anchor." As 


certain asPyrgi signifies "towers," so certainly does i?o»20 signify 
" strength ^" and I believe that those are quite right who con- 
sider that the name Eoma in this sense is not accidental. 
This Eoma is described as a Pelasgian place in which Evander, 
the introducer of scientific culture resided. According to tra- 
dition, the first foundation of civilisation was laid by Saturn, 
in the golden age of mankind. The tradition in Virgil, who 
was extremely learned in matters of antiquity, that the first 
men were created out of trees must be taken quite literally*; 
for as in Greece the fMvpfir]Ke<i were metamorphosed into the 
Myrmidons, and the stones thrown by DeucaHon and Pyrrha 
into men and women, so in Italy trees, by some divine power, 
were changed into human beings. These beings, at first only 
half human, gradually acquired a civilisation which they owed 
to Saturn; but the real intellectual culture was traced to 
Evander, who must not be regarded as a person who had come 
from Arcadia, but as the good man, as the teacher of the alpha- 
bet and of mental culture, which man gradually works out for 

The Eomans clung to the conviction that Eomulus, the 
founder of Eome, was the son of a virgin by a god, that his 
life was marvellously preserved, that he was saved from the 
floods of the river and was reared by a she-wolf. That this 
poetry is very ancient, cannot be doubted ; but did the legend 
at all times describe Eomulus as the son of Eea Silvia or Ilia? 
Perizonius was the first who remarked against Ej^ckius, that 
Eea Ilia never occurs together, and that Eea Silvia was a 
daughter of Numitor, while Ilia is called a daughter of Aeneas. 
He is perfectly right: Naevius and Ennius called Eomulus a 
son of Ilia, the daughter of Aeneas, as is attested by Servius 
on Virgil and Porphyrio on Horace^; but it cannot be hence 

' It is well known that there is in Stobaeus (vii. 13) a poem upon Rome, 
which is ascribed to Erinna. But as Erinna composed her poems at a time 
when Rome cannot be supposed to have been reno-«Tied in Aeolia, commentators 
have imagined the poem to be a hymn on Strength. But Strength cannot be 
called a daughter of Ares; Strength might rather be said to be his mother. The 
poem belongs to a much later date, and proceeding on this supposition it may 
perh.ips be possililc for some one to discover the real name of the author. It 
certainly belongs to the period subsequent to the Hannibalian war, and was 
perhaps not written till the time of the emperors; but to me it seems most pro- 
bable that the author was a contemporary of Sulla. — N. 

* Gensque virum truncis et duro robore nata. Virgil, Aen. viii, 315. 

* Carm. i. 2. 


inferred, that tliis was the national opinion of the Romans 
themselves, for the poets who were familiar with the Greeks, 
might accommodate their stories to Greek poems. The ancient 
Eomans, on the other hand, could not possibly look upon the 
mother of the founder of their city as a daughter of Aeneas, 
who was believed to have lived 333 or 360 years earlier. 
Dionysius says that his account, which is that of Fabius, occur- 
red in the sacred songs, and it is in itself perfectly consistent. 
Fabius cannot have taken it, as Plutarch asserts, from Diodes, 
a miserable unknown Greek author; the statue of the she- 
wolf was erected in the year A. U. 457, long before Diodes 
wrote, and at least a hundred years before Fabius. This tra- 
dition therefore is certainly the more ancient Roman one ; and 
it puts Rome in connection with Alba. A monument has 
lately been discovered at Bovillae : it is an altar which the 
Gentiles Julii erected lege Albana, and therefore expresses a 
relio-ious relation of a Roman f?ens to Alba. The connetion 
of the two towns continues down to the founder of Rome; 
and the well known tradition, with its ancient poetical details, 
many of which Livy and Dionysius omitted from their histories 
lest they should seem to deal too much in the marvellous, runs 
as follows. 

Numitor and Amulius were contending for the throne of 
Alba.^ Amulius took possession of the throne, and made Rea 
Silvia, the daughter of Xumitor, a vestal virgin, in order that 
the Silvian house might become extinct. This part of the story 
was composed without any insight into political lawSj for a 
daughter could not have transmitted any gentilician rights. 
The name Rea Silvia is ancient, but Rea is only a surname : 
rea femmina often occurs in Boccaccio, and is used to this day 
in Tuscany to designate a woman whose repiitation is blighted: 
a priestess Rea is described by Virgil as having been over- 
powered by Hercules. While Rea was fetching water in a 
grove for a sacrifice the sun became eclipsed, and she took 
refuge from a wolf in a cave where she was overpowered by 
Mars. When she was delivered, the sun was again eclipsed 
and the statue of Yesta covered its eyes. Livy has here 
abandoned the marvellous. The tyrant threw Rea with her 

* Numitor is a praenomen, Imt the name Amulius docs not shew that he 
belonged to the gens Silvia: I therefore doubt whctlicr tlie ancient tradition 
represented them as brothers. — N. 


infcints into the river Anio : she lost her life in the waves, but 
the god of the river took her soul and changed it into an 
immortal goddess whom he married. This story has been 
softened down into the tale of her imprisonment, which is un- 
poetical enough to be a later invention. The river Anio carried 
the cradle like a boat into the Tiber, and the latter conveyed 
it to the foot of the Palatine, the water having overflowed the 
country, and the cradle was upset at the root of a fig-tree. A 
she- wolf carried the babes away and suckled them 7; ]\lars sent 
a woodpecker which provided the children with food, and the 
bird parra^ which protected them from insects. These state- 
ments are gathered from various quarters; for the historians 
got rid of the marvellous as much as possible. Faustulus, the 
legend continues, found the boys feeding on the milk of the 
huge wild-beast, he brought them up with his twelve sons, 
and they became the staunchest of all. Being at the head of 
the shepherds on Mount Palatine, they became involved in a 
quarrel with the shepherds of Numitor on tlie Aventine — the 
Palatine and the Aventine are always hostile to each other — 
Eemus being taken prisoner was led to Alba, but Eomulus 
rescued him, and their descent from Numitor being discovered, 
the latter was restored to the throne, and the two young men 
obtained permission to form a settlement at the foot of Mount 
Palatine where they had been saved. 

Out of this beautiful poem, the falsifiers endeavoured to 
make some credible story: even the unprejudiced and poetical 
Livy tried to avoid the most marvellous points as much as he 
could, but the falsifiers went a step farther. In the days when 
men had altogether ceased to believe in the ancient gods, 
attempts were made to find something intelligible in the old 
legends, and thus a history was made up, which Plutarch fondly 
embraced and Dionysius did not reject, though he also relates 
the ancient tradition in a mutilated form. He says that many 
people believed in daemons, and that such a daemon might 
have been the father of Eomulus; but he himself is very far 
from believing it, and rather thinks that Amulius himself, in 
disguise, violated Eca Silvia amid thunder and lightning 
produced by artifice. Tliis he is said to have done in order to 
have a pretext for getting rid of her, but being entreated by 

' In Eastern legends, children are nourished with theman-ow of lions.— N. 
* Scrv. on Virg. Aen. i. 274. 


Ills daughter not to drown her, he hiiprisoned her for life. The 
children were saved by the shepherd, who was commissioned 
to expose them, at the request of Numitor, and two other boys 
were put in their place. Numitor's grandsons were taken to 
a friend at Gabii, who caused them to be educated according 
to their rank and to be instructed in Greek literature. Attempts 
have actually been made to introduce this stupid forgery into 
history, and some portions of it have been adopted in the nar- 
rative of our historians; for example, that the ancient Alban 
nobility migrated with the two brothers to Rome; but if this 
had been the case there would have been no need of opening 
an asylum, nor would it have been necessary to obtain by force 
the connuhium with other nations. 

But of more historical importance is the difference of 
opinion between the two brothers, respecting the building of 
the city and its site. According to the ancient tradition, both 
were kings and the equal heads of the colony ; Romulus is 
universally said to have wished to build on the Palatine, while 
Remus, according to some, preferred the Aventinc ; according 
to others, the hill Remuria. Plutarch states that the latter is 
a hill three miles south of Rome, and cannot have been any 
other than the hill nearly opposite St. Paul, which is the more 
credible, since this hill, though situated in an otherwise un- 
healthy district, has an extremely fine air: a very important 
point in investigations respecting the ancient Latin towns, for 
it may be taken for certain, that where the air is now healthy 
it was so in those times also, and that where it is now decidedly 
unhealthy, it was anciently no better. The legend now goes 
on to say, that a dispute arose between Romulus and Remus as 
to which of them should give the name to the town, and also 
as to where it was to be built. A town Remuria therefore 
undoubtedly existed on that hill, though subsequently we find 
the name transferred to the Aventinc, as is the case so frequently. 
Accordins: to the common tradition auoiirs were to decide 
between the brothers; Romulus took his stand on the Palatine, 
Remus on the Aventinc. The latter observed the whole night 
but saw nothing until about sunrise, when he saw six vultures 
flying from north to south and sent word of it to Romulus, but 
at that very time the latter, annoyed at not having seen any 
sign, fraudulently sent a messenger to say that he had seen 
twelve vultures, and at the very moment the messenger airivod, 


tliere did appear twelve vultures, to which Romulus appealed. 
This account is impossible; for the Palatine and Aventine are 
so near eacli other that, as every Roman well knew, whatever 
a person on one of the two hills saw high in the air, could not 
escape the observation of any one who was watching on the 
other. This part of the story therefore cannot be ancient, and 
can be saved only by substituting the Remuria for the Aventine. 
As the Palatine was the seat of the noblest patrician tribe, and 
the Aventine the special town of the plebeians, there existed 
between the two a perpetual feud, and thus it came to pass that 
in after times the story relating to the Remuria, which was 
far away from the city, was transferred to tlie Aventine. 
According to Ennius, Romulus made his observations on the 
Aventine ; in this case Remus must certainly have been on the 
Remuria, and it is said that when Romulus obtained the augury 
he threw his spear towards the Palatine. This is the ancient 
legend which was neglected by the later writers. Romulus 
took possession of the Palatine. The spear taking root and 
becoming a tree, which existed down to the time of Nero, is a 
symbol of the eternity of the new city, and of the protection 
of the gods. The statement that Romulus tried to deceive his 
brother is a later addition; and the beautiful poem of Ennius 
quoted by CiceroS knows nothing of this circumstance. The 
conclusion Avhicli must be drawn from all this is, that in the 
earliest times there were two towns, Roma and Remuria, the 
latter being far distant from the city and from the Palatine. 

Romulus now fixed the boundary of his town, but Remus 
scornfully leapt across the ditch, for which he was slain by 
Celer, a hint that no one should cross the fortifications of Rome 
with impunity. But Romulus fell into a state of melancholy 
occasioned by the death of Remus; he instituted festivals to 
honor him and ordered an empty throne to be put up by the 
side of his own. Thus we have a double kingdom which ends 
with the defeat of Remuria. 

The question now is what were these two towns of Roma 
and Remuria? They were evidently Pelasgian places; the 
ancient tradition states that Sicelus migrated from Rome south- 
ward to the Pelasgians, that is, the Tyrrhenian Pelasgians were 
pushed forAvard to the IMorgetes, a kindred nation in Lucania 
and in Sicily. Among the Greeks it was, as Dionysius states, 

» De Divinat. i. 48. 


a general opinion, that Rome was a Pelasgian, that is a Tyr- 
rhenian city, but the authorities from whom he learnt this are 
no longer extant. There is, however, a fragment in which it is 
stated that Rome was a sister city of Antium and Ardea. Here 
too we must apply the statement from the chronicle of Cumae, 
that Evander, Avho, as an Arcadian, was likewise a Pelasgian, 
had his palatium on the Palatine. To us he appears of less 
importance than in the legend, for in the latter he is one of 
the benefactors of nations, and introduced among the Pelas- 
gians in Italy the use of the alphabet and other arts, just as 
Damaratus did among the Tyrrhenians in Etruria. In this 
sense^ therefore, Rome was certainly a Latin town, and had 
not a mixed but a purely Tyrrhene- Pelasgian population. The 
subsequent vicissitudes of this settlement may be gathered from 
the allegories. 

Romulus now found the number of his fellow-settlers too 
small; the number of 3000 foot and 300 horse, which Livy 
gives from the commentaries of the pontiffs, is worth nothing; 
for it is only an outline of the later military an^angement trans- 
ferred to the earliest times. According to the ancient tradition, 
Romulus's band was too small, and he opened an asylum on 
the Capitoline hill. This asylum, the old description states, 
contained only a very small space, a proof how little these 
things were understood historically. All manner of people, 
thieves, murderers, and vagabonds of every kind flocked thither. 
This is the simple view taken of the origin of the clients. In 
the bitterness with which the estates subsequently looked upon 
one another, it was made a matter of reproach to the Patricians, 
that their earliest ancestors had been vagabonds; though it was 
a common opinion, that the patricians were descended from 
the free companions of Romulus, and that those who took 
refuge in the asylum placed themselves as clients under the 
protection of the real free citizens. But now they wanted 
women, and attempts were made to obtain the commbium Avitli 
neighbouring towns, especially perhaps with Antemnac, which 
Avas only four miles distant from Rome, with the Sablncs and 
others. This being refused, Romulus had recourse to a stra- 
tagem, proclaiming that he had discovered the altar of Consus, 
the god of counsels, an allegory of his cunning in general. In 
the midst of the solemnities, the Sabine maidens, thirty in 
number, were carried off, from whom the curiae received their 

VOL. I. . D 


names: tliis is tlie genuine ancient legend, and it proves Kow 
small ancient Rome was conceived to have been. In later 
times tlie number was thouglit too small, it was supposed tliat 
these thirty had been chosen by lot foj the purpose of naming 
the curiae after them ; and Valerius Antias fixed the number of 
the women who had been carried off at five hundred and twenty- 
seven. The rape is placed in the fourth month of the city, 
because the consualia fall in August, and the festival comme- 
morating the foundation of the city in April; later writers, as 
Cn. Gellius, extended this period to four years, and Dionysius 
found this of course far more credible. From this rape there 
arose wars, first with the neiglibouring towns which were de- 
feated one after another, and at last with the Sabines. The 
ancient legend contains not a trace of this war having been of 
long continuance ; but in later times it was necessarily supposed 
to have lasted for a considerable time, since matters were then 
measured by a different standard. Lucumo and Caelius came 
to the assistance of Romulus, an allusion to the expedition of 
Caeles Vibenna, which however belongs to a much later period. 
The Sabine king, Tatius, was induced by treachery to settle 
on the hill which is called the Tarpeian arx. Between the 
Palatine and the Tarpeian rock a battle was fought, in which 
neither party gained a decisive victory, until the Sabine women 
threw themselves between the combatants, who agreed that 
henceforth the sovereignty should be divided between the Ro- 
mans and Sabines. According to the annals, this happened in 
the fourth year of Rome. 

But this arrangement lasted only a short time ; Tatius was 
slain during a sacrifice at Lavinium , and his vacant throne was 
not filled up. During their common reign, each king had a 
senate of one hundred members, and the two senates, after con- 
sulting separately, used to meet, and this was called comitium. 
Romulus durincv the remainder of his life ruled alone; the 
ancient legend knows nothing of his having been a tyrant: 
according to Ennius he continued, on the contrary, to be a 
mild and benevolent king, while Tatius was a tyrant. The 
ancient tradition contained nothing beyond the beginning and 
the end of the reigji of Romulus; all that lies between these 
points, the war with the Veien tines, Fidenates, and so on, is a 
foolish invention of later annalists. The poem itself is beautiful, 
but this inserted narrative is highly absurd, as for example the 



statement that Eomulus slew 10,000 Veientines with his own 
hand. The ancient poem passed on at once to the time when 
RomuUis had completed his earthly career, and Jupiter fulfilled 
his promise to j\Iars, that Romulus was the only man whom he 
would introduce among the gods. According to this ancient 
legend, the king was reviewing his army near the marsh of 
Caprae^ when, as at the moment of his conception, there oc- 
curred an eclipse of the sun and at the same time a hurricane, 
during which INIars descended in a fiery chariot and took his 
son up to heaven. Out of this beautifid poem the most 
Avretchcd stories have been manufactured; Romulus, it is said, 
while in the midst of his senators was knocked down, cut into 
pieces, and thus carried away by them under their togas. This 
stupid story was generally adopted, and that a cause for so 
horrible a deed might not be wanting, it was related that in 
his latter years Romulus had become a tyrant, and that the 
senators took revenge by murdering him. 

After the death of Romulus, the Romans and the people of 
Tatius quarrelled for a long time with each other, the Sabines 
wishing that one of their nation should be raised to the throne, 
while the Romans claimed that the new king should be chosen 
from among them. At length they agreed, it is said, that the 
one nation should choose a king from the other. 

We have now reached the point at which it is necessary to 
speak of the relation between the two nations, such as it 
actually existed. 

All the nations of antiquity lived in fixed forms, and their 
civil relations were always marked by various divisions and 
sub-divisions. When cities raise themselves to the rank of 
nations, we alwaj^s find a division at first into tribes; Herodotus 
mentions such tribes in the colonisation of Cyrene, and the 
same was afterwards the case at the foundation of Thurii ; but 
when a place existed anywhere as a distinct township, its nature 
was characterised bv the fact of its citizens beinc; at a certain 
time divided into gentes (jevi]) each of which had a common 
chapel and a common hero. These gentes were united in defi- 
nite numerical proportions into curiae {(ppdrpai). The gentes 
are not families but free corporations, sometimes close and 
sometimes open; in certain cases, the whole body of the state 
might assign to them new associates; the great coimcil at 
Venice was a close body, and no one could be admitted whose 



ancestors had not been in it, and sucli also was tlie case in 
many oligarchical states of antiquity. 

All civil communities had a council and an assembly of 
burghers, that is a small and a great council; the burghers 
consisted of the guilds or gentes, and these again were united, 
as it were, in parishes; all the Latin towns had a council of 100 
members, who were divided into ten curiae ; this division gave 
rise to the name o^ decuriones, which remained in use as a title 
of civic magistrates down to the latest times, and through the 
lex Julia was transferred to the constitution of the Italian 
municipia. That this council consisted of one hundred persons 
has been proved by Savigny, in the first volume of his history 
of the Koman law. This constitution continued to exist till a 
late period of the middle ages, but perished when the institu- 
tion of guilds took the place of municipal constitutions. 
Giovanni Villani says, that previously to the revolution in the 
twelfth century there were at Florence 100 buoni uomini, who 
had the administration of the city. There is nothing in our 
German cities which answers to this constitution. We must 
not conceive those hundred to have been nobles ; they were 
an assembly of burghers and country people, as was the case 
in our small imperial cities, or as in the small cantons of 
Switzerland. Each of them represented a gens ; and they are 
those whom Propertius calls patres pelliti. The curia of Rome, 
a cottage covered with straw ^", was a faithful memorial of the 
times when Kome stood buried in the night of history, as a 
small country town surrounded by its little domain. 

The most ancient occurrence which we can discover from 
the form of the allegory, by a comparison of what happened in 
other parts of Italy, is a result of the great and continued com- 
motion among the nations of Italy. It did not terminate when 
the Oscans had been pressed forward from lake Fucinus to the 
lake of Alba, but continued much longer. The Sabines may 
have rested for a time, but they advanced far beyond the districts 
about which we have any traditions. These Sabines began 
as a very small tribe, but afterwards became one of the greatest 
nations of Italy, for the Marrucinians, Caudines, Vestinians, 
Marsians, Pclignians, and in short all the Samnite tribes, the 
Lucanians, the Oscan part of the Bruttians, the Picentians and 
several others were all descended from the Sabine stock, and 
'" liecens horrcbat regia culmo. Virgil, 


yet there are no traditions about their settlements except in a 
few cases. At the time to which we must refer the foundation 
of Rome, the Sabines were widely diffused. It is said that, 
guided by a bull, they penetrated into Opica, and thus occu- 
pied the country of the Samnites. It was perhaps at an earlier 
time that they migrated down the Tiber, whence we there find 
Sabine towns mixed with Latin ones; some of their places also 
existed on the Anio. The country afterwards inhabited by 
the Sabines was probably not occupied by them till a later 
period, for Falerii is a Tuscan town, and its population was 
certainly at one time thoroughly Tyrrhenian. 

As the Sabines advanced, some Latin towns maintained 
their independence, others were subdvied; Fidenae belonged 
to the former, but north of it all the country was Sabine. Now 
by the side of the ancient Roma we find a Sabine town 
on the Quirinal and Capitoline close to the Latin town; but 
its existence is all that we know about it. A tradition states, 
that there previously existed on the Capitoline a Siculian 
town of the name of Saturnia^^ which, in this case, must have 
been conquered by the Sabines. But whatever we may think 
of this, as well as of the existence of another ancient town on 
the Janiculum, it is certain that there were a number of small 
towns in that district. The two towns could exist perfectly 
well side by side, as there was a deep marsh between them. 

The town on the Palatine may for a long time have been in 
a state of dependence on the Sabine conqueror whom tradition 
calls Titus Tatius; hence he was slain during the Laurentinc 
sacrifice, and hence also his memoiy was hateful.^" The exist- 
ence of a Sabine town on the Quirinal is attested by the un- 
doubted occurrence there of a number of Sabine chapels, which 
were known as late as the time of Yarro, and from which he 
proved that the Sabine ritual was adopted by the Romans. 
This Sabine element in the Avorship of the Romans has almost 
always been overlooked^-', in consequence of the prevailing 

" Varro, L.L. v. (iv.) 42. 

'- Eniiius calls him a tyrant in the well -known verse: O Tite, tiite, Tati, tibi 
tanta ti/i-anni tiilisti — N. 

'^ I have spent many days at Rome in searching after the ancient churchcs,whieh 
were pulled down at the time when the town was splendidly rebuilt; hut I never 
was able to see my way, until I read the work of a priest of the seventeenth 
century, who pointed out the traces of them which .still exist; .ind I conceive 
that it was in a similar manner tliat Yarro jioinled out the sites of the Sabine 
chapels and nacella on the Quirinal and Capitoline. — N. 


desire to look upon every thing as Etruscan; but, I repeat, 
there is no doubt of the Sabine settlement, and that it was the 
result of a great commotion among the tribes of middle Italy. 


The tradition that the Sabine women were carried off, because 
there existed no connubium, and that the rape was followed by a 
war, is undoubtedly a symbolical representation of the relation 
between the two towns, previous to the establishment of the 
right of intermarriage ; the Sabines had the ascendancy and 
refused that right, but the Romans gained it by force of arms. 
There can be no doubt, that the Sabines were originally the 
ruling people, but, that in some insurrection of the Romans 
various Sabine places, such as Antemnae, Fidenae and others, 
were subdued, and thus these Sabines were separated from 
their kinsmen. The Romans therefore re-established their 
independence by a war, the result of which may have been 
such as we read it in the tradition — Romulus being, of course, 
set aside — namely that both places as two closely united towns 
formed a kind of confederacy, each with a senate of 100 mem- 
bers, a king, an offensive and defensive alliance, and on the 
understanding that in common deliberations the burghers of 
each should meet together in the space between the two towns 
which was afterwards called the comitium. In this manner 
they formed a united state in regard to foreign nations. 

The idea of a double state was not unknown to the ancient 
writers themselves, although the indications of it are preserved 
only in scattered passages, especially in the scholiasts. The 
head of Janus, which in the earliest times was represented on 
the Roman as, is the symbol of it, as has been correctly observed 
by writers on Roman antiquities. The vacant throne by the 
side of the curule chair of Romulus points to the time when 
there was only one king, and represents the equal but quiescent 
right of the other people.^ 

That concord was not of long duration is an historical fact 

' Comp. above, page 40. 


likewise ; nor can it be doubted that the Eoman king assumed 
the supremacy over the Sabines, and that in consequence the 
two councils were united so as to form one senate under one 
king, it being agreed that the king should be alternately a 
Roman and a Sabine, and that each time he should be chosen 
by the other people : the king, however, if displeasing to the 
non-electing people, was not to be forced upon them, but was 
to be invested with the impfrium only on condition of the 
auguries being favourable to him, and of his being sanctioned 
by the whole nation. The non-electing tribe accordingly had 
the right of either sanctioning or rejecting his election. In the 
case of Nuraa this is related as a feet, but it is only a disguise- 
ment of the right derived from the ritual books. In this 
manner the strange double election, which is otherwise so 
mysterious and was formerly completely misunderstood, be- 
comes quite intelligible. One portion of the nation elected and 
the other sanctioned; it being intended that, for example, the 
Romans should not elect from among the Sabines a king 
devoted exclusively to their own interests, but one who was at 
the same time acceptable to the Sabines. 

When, perhaps after several generations of a separate 
existence, the two states became united, the towns ceased to be 
towns, and the collective body of the burghers of each became 
tribes, so that the nation consisted of two tribes. The form of 
addressing the Roman people was from the earliest times 
Populus Romanus Quirites, which, when its origin was forgotten, 
was changed into Populus Romanus Quiritium, just as lis vin- 
diciae was afterwards chanoed into lis vindiciarum. This chansi'e 
is more ancient tlian Llvy ; the correct expression still continued 
to be used, but was to a great extent supplanted by the false 
one. The ancient tradition relates that after the union of the 
two tribes the name Quirites was adopted as the common 
designation for the whole people; but this is erroneous, for the 
name was not used in this sense till a very late period. This 
designation remained in use and was transferred to the plebeians 
at a time when the distinction between Romans and Sabines, 
between these tw^o anel the Liiccrcs, na}', when even th;it 
between patricians and plebeians had almost ceased to be 
noticed.^ Thus the two towns stood side by side as tribes 

■^ This is not my discovery : it belongs to the great prcsiileiit of the French 
Parliament, Barnabas Brissouiu?, from whom \\c may still learn mucli, although 


forming one state, and it is merely a recognition of the ancient 
tradition when we call the Latins Ranmes, and the Sabines 
Titles : that the derivation of these appellations from Eomiilus 
and T. Tatlus is incorrect is no argument against the view 
here taken. 

Dion3^sius, who had good materials and made use of a great 
many, must, as far as the consular period is concerned, have 
had more than he gives ; there is in particular one important 
change in the constitution, concerning which he has only a 
few words, either because he did not see clearly or because he 
was careless.^ But as regards the kingly period, he was well 
acquainted with his subject; he says that there was a dispute 
between the two tribes respecting the senates, and that Numa 
settled it, by not depriving the Ramnes, as the first tribe, of 
any thing, and by conferring honours on the Titles. This is 
perfectly clear. The senate, which had at first consisted of 100 
and now of 200 members, was divided into ten decuries, each 
being headed by one, who was its leader; these are the decern 
primi, and they were taken from the Ramnes. They formed 
the college, which, when there was no king, undertook the 
government one after another, each for five days, but in such 
a manner, that they ahvays succeeded one another in the same 
order, as we must believe with Livy, for Dionysius here 
introduces his Greek notions of the Attic prytanes, and Plu- 
tarch misunderstands the matter altogether. 

After the example of the senate the number of the augurs 
and pontiffs also was doubled, so that each college consisted of 
four members, tAvo being taken from the Ramnes and two from 
the Titles. Although it is not possible to fix these changes 
chronologically, as Dionysius and Cicero do, yet they are as 
historically certain as if we actually knew the kings who intro- 
duced them. 

Such was Rome in the second stage of its development. 

we may correct a gi'cat many trifling errors of detail into which he fell; but 
where sliould we now be, had there not been such men as Brissonius, Scaliger, 
and Cujacius. Brissonius however on the point here in question goes too far; 
for he wishes to emend every where: all exaggeration injures trutli, and the 
consequence was, that many persons altogether refused to follow him because 
he often erred. The ingenious J. F. Gronovius opposed him, and referred to 
passages in Livy which were against him; but, as was remarked above, the 
erroneous expression was established previously to the time of Livy. — N. 
' See Hist. Rom. vol. ii. p. 179, 220, etc. 


This period of equalisation is one of peace, and is described as 
tlae reign of Xunia about whom the traditions are simple and 
brief. It is the picture of a peaceful condition with a holy 
man at the head of affairs, like Nicolas von der Flue in Switzer- 
land. Numa was supposed to have been inspired by the 
goddess Egeria, to whom he was married in the grove of the 
Camenae, and who introduced him into the choir of her sisters ; 
she melted away in tears at his death, and thus gave her name 
to the spring which arose out of her tears. Such a peace of 
forty years, during which no nation rose against Eome because 
Xuma's piety was communicated to the surrounding nations, 
is a beautiful idea, but historically impossible in those times, 
and manifestly a poetical fiction. 

The death of Kuma forms the conclusion of the first saeculum, 
and an entirely new period follows, just as in the Theogony of 
Hesiod the age of heroes is followed by the iron age; there is 
evidently a change, and an entirely new order of things is con- 
ceived to have arisen. Up to this point we have had nothing 
except poetry, but with Tullus Hostilius a kind of history 
begins, that is, events are related which must be taken in 
general as historical, though in the light in which they are 
presented to us they are not historical. Thus, for example, 
the destruction of Alba is historical, and so in all probability 
is the reception of the Albans at Rome. The conquests of 
Ancus Martins are quite credible; and they appear like an oasis 
of real history in the midst of fables. A similar case occurs 
once in the chronicle of Cologne. In the Abyssinian annals, 
we find in the thirteenth century a very minute account of 
one particular event, in which we recognise a piece of contem- 
poraneous history, though we meet with nothing historical 
either before or after. 

The history which then follows is like a picture viewed from 
the wrong side, like phantasmata; the names of the kings are 
perfectly fictitious; no man can tell how long the Roman 
kings reigned, as we do not know how many there were, since 
it is only for the sake of the number, that seven were supposed 
to have ruled, seven being a number which appears in many 
relations, especially in important astronomical ones. Hence 
the chronological statements are utterly worthless. We must 
conceive as a succession of centuries, the period from the origin 
of Rome down to the times wherein were constructed the 


enormous works, such as tlie great drains, tlie wall of Servius 
and others, which were actually executed under the kings, and 
rival the great architectural works of the Egyptians. Itomu- 
lus and Numa must be entirely set aside; but a long period 
follows, in which the nations gradually unite and develop 
themselves until the kingly government disappears and makes 
way for republican institutions. 

But it is nevertheless necessary to relate the history, such as 
it has been handed down, because much depends upon it. 
There was not the slightest connection between Rome and 
Alba, nor is it even mentioned by the historians, though they 
suppose that Rome received its first inhabitants from Alba; but 
in the reign of Tullus Hostilius the tw^o cities on a sudden ap- 
pear as enemies: each of the two nations seeks war, and tries 
to allure fortune by representing itself as the injured party, 
each wishing to declare war. Both sent ambassadors to demand 
reparation for robberies which had been committed. The form 
of procedure was this: the ambassadors, that is the Fetiales, 
related the grievances of their city to every person they met, 
they then proclaimed them in the market place of the other 
city, and if, after the expiration of thrice ten days no repara- 
tion was made, they said: "We have done enough and now 
return," whereupon the elders at home held counsel as to how 
they should obtain redress. In this formula accordingly the 
res, that is the surrender of the guilty and the restoration of 
the stolen property must have been demanded. Xow it is 
related that the two nations sent such ambassadors quite simul- 
taneously, but that Tullus Hostilius retained the Alban am- 
bassadors, ^^ntil he was certain that the Romans at Alba had 
not obtained the justice due to them, and had therefore de- 
clared war. After this he admitted the ambassadors into the 
senate, and the reply made to their complaint was, that they 
themselves had not satisfied the demands of the Romans. Livy 
then continues: belkim in trigesimum diem dixerant. But the 
real formula is, post trigesimum diem, and we may ask. Why 
did Livy or the annalist whom he followed make this alteration? 
For an obvious reason: a person may ride from Rome to Alba 
in a couple of hours, so that the detention of the Alban am- 
bassadors at Rome for thirty days, without their hearing what 
was going on in the mean time at Alba, was a matter of im- 
possibility: Livy saw this, and therefore altered the formula. 


But the ancient poet was not concerned about such things, and 
without hesitation increased the distance in his imagination, 
and represented Koine and Alba as great states. 

The whole description of the circumstances under which the 
fate of Alba was decided is just as manifestly poetical, but we 
shall dwell upon it for a while in order to show how a sem- 
blance of history may arise. Between Rome and Alba there 
was a ditch, Fossa Cluilia or Cloelia, and there must have been 
a tradition that the Albans had been encamped there; Livy 
and Dionysius mention that Cluilius, a general of the Albans, 
had given the ditch its name, having perished there. It was 
necessary to mention the latter circumstance, in order to ex- 
plain the fact that afterwards their general was a different 
person, Mettius Fuffetlus, and yet to be able to connect the 
name of that ditch with the Albans. The two states committed 
the decision of their dispute to champions^ and Dionysius says, 
that tradition did not agree as to whether the name of the 
Eoman champions was Horatii or Curiatii, although he him- 
self, as well as Livy, assumes that it was Horatii, probably 
because it was thus stated by the majority of the annalists. 
"Who would suspect any uncertainty here if it were not for this 
passage of Dionysius? The contest of the three brothers on 
each side is a symbolical indication that each of the two states 
was then divided into three tribes. Attempts have indeed 
been made to deny that the three men were brothers of the 
same birth, and thus to remove the improbability; but the 
legend went even further, representing the three brothers on 
each side as the sons of two sisters, and as born on the same 
day. This contains the suggestion of a perfect equality be- 
tween Eome and Alba. The contest ended in the complete 
submission of Alba; it did not remain faithful, however, and 
in the ensuing struggle with the Etruscans, !Mcttius Fuffetlus 
acted the part of a traitor towards Rome, but not being able 
to carry his design into effect, he afterwards fell upon the fugi- 
tive Etruscans. Tullus ordered him to be torn to pieces and 
Alba to be razed to the ground, the noblest Alban lamilies 
being transplanted to Rome. The death of Tullus is no less 
poetical. Like Numa he undertook to call down lightning 
from heaven, but he thereby destroyed himself and his house. 

If we endeavour to discover the historical substance of these 
legends, we at once find ourselves in a period when Rome no 


longer stood alone, but had colonies with Koman settlers, pos- 
sessing a third of the territory and exercising sovereign power 
over the original inhabitants. This was the case in a small 
number of towns for the most part of ancient Siculian origin. 
It is an undoubted fact that Alba was destroyed, and that after 
this event the towns of the Prisci Latini formed an inde- 
pendent and compact confederacy; but whether Alba fell in 
the manner described, whether it was ever compelled to recog- 
nise the supremacy of Rome, and whether it was destroyed 
by the Romans and Latins conjointly, or by the Romans or 
Latins alone, are questions which no human ingenuity can 
solve. It is however most probable, that the destruction 
of Alba was the work of the Latins, who rose ao-ainst her 
supremacy: whether in this case the Romans received the 
Albans among themselves, and thus became their benefac- 
tors instead of destroyers, must ever remain a matter of uncer- 
tainty. That Alban families were transplanted to Rome cannot 
be doubted, any more than that the Prisci Latini from that 
time constituted a compact state; if we consider that Alba 
was situated in the midst of the Latin districts, that the Alban 
mount Avas their common sanctuary, and that the grove of 
Ferentina was the place of assembly for all the Latins, it 
must appear more probable that Rome did not destroy Alba, 
but that it perished in an insurrection of the Latin towns, and 
that the Romans strengthened themselves by receiving the 
Albans into their city. 

Whether the Albans were the first that settled on the Caelian 
hill, or whether it was previously occupied cannot be decided. 
The account which places the foundation of the town on the 
Caelius in the reign of Romulus suggests that a town existed 
there before the reception of the Albans; but what is the au- 
thenticity of this account? A third tradition represents it as 
an Etruscan settlement of Caeles Vibenna. Thus much is cer- 
tain that the destruction of Alba greatly contributed to increase 
the power of Rome. There can be no doubt that a third town 
whicli seems to have been very populous, now existed on the 
Caelius and on a portion of the Esquiliae : such a settlement 
close to other towns was made for the sake of mutual protec- 
tion. Between the two more ancient towns there continued to 
be a marsh or swamp, and Rome was protected on the south 
by stagnant water; but between Rome and the third town 


tliere was a dry plain. Eome also had a considerable suburb 
towards the Aventine protected by a wall and a ditch, as is 
implied in the story of Kemus. He is a personification of the 
plebs, leaping across the ditch from the side of the Aventine, 
though we ought to be very cautious in regard to allegory. 

The most ancient town on the Palatine was Rome; the 
Sabine town also must have had a name, and I have no doubt 
that, according to common analogy, it was Quirium, the name 
of its citizens being Quirites. This I look upon as certain. I 
have almost as little doubt that the town on the Caelian was 
called Lucerum, because when it was united with Rome, its citi- 
zens were called Lucertes (Luceres). The ancients derive this 
name from Lucumo, king of the Tuscans, or from Lucerus, king 
of Ardea; the latter derivation probably meaning, that the race 
was Tyrrheno-Latin, because Ardea was the capital of that 
race. Rome was thvis enlarged by a thii'd element, which, 
however, did not stand on a footing of equality with the two 
others, but was in a state of dependance similar to that of Ire- 
land relatively to Great Britain down to the year 1782. But 
although the Luceres were obliged to recognise the supremacy 
of the two elder tribes, they were considered as an integral 
part of the whole state, that is, as a third tribe with an admin- 
istration of its own, but inferior riirhts. A\liat throws lioht 
upon our way here, is a passage of Festus who is a great autho- 
rity on mattei's of Roman antiquity, because he made his ex- 
cerpts from Yerrius Flaccus ; it is only in a few points that, in 
my opinion, either of them was mistaken ; all the rest of the 
mistakes in Festus may be accounted for by the imperfection 
of the abridgment, Festus not always understanding Vcrrius 
Flaccus. The statement of Festus to which I here allude, is, 
that Tarquinius Siiperbus increased the number of the Vestals, 
in order that each tribe miirht have two. With this we must 
connect a passage from the tenth book of Livv, where he says 
that the augurs were to represent the three tribes. The num- 
bers in the Roman colleges of priests Avere always multiples 
either of two or of three ; the latter was the case with the Vestal 
Virgins and the great Flamincs, and the former with the Au- 
gurs, Pontifls and Fetiales, who represented only the first two 
tribes. Previously to the passing of the Ogulnian law the 
number of augurs was four, and wlien subsequently five ple- 
beians were added, the basis of this increase was difiercnt, it is 


true*, but the ancient rule of the number being a multiple of 
three was preserved. The number of pontiffs which was then 
four, was increased only by four: this might seem to contra- 
dict what has just been stated, but it has been overlooked that 
Cicero speaks of five new ones having been added, for he in- 
cluded the Pontifex Maximus, which Livy does not. In like 
manner there were twenty Fetiales, ten for each tribe. To the 
Salii on the Palatine, Numa added another brotherhood on the 
Quirinal; thus we everywhere see a manifest distinction 
between the first two tribes and the third, the latter being 
treated as inferior. 

The third tribe, then, consisted of free citizens, but they had 
not the same rights as the members of the first two; yet its 
members considered themselves superior to all other people ; 
and their relation to the other two tribes was the same as that 
existing between the Venetian citizens of the main land and the 
nobili. A Venetian nobleman treated those citizens with far 
more condescension than he displayed towards others, provided 
they did not presume to exercise any authority in political mat- 
ters. Whoever belonged to the Luceres, called himself a 
Roman, and if the very dictator of Tusculum had come to 
Rome, a man of the third tribe there would have looked upon 
him as an inferior person, though he himself had no influence 

Tullus was succeeded by Ancus. Tullus appears as one of 
the Ramnes, and as descended from Hostus Hostilius, one of 
the companions of Romulus ; but Ancus was a Sabine, a grand- 
son of Numa. The accounts about him are to some extent 
historical, and there is no trace of poetry in them. In his 
reign, the development of the state again made a step in ad- 
vance. According to the ancient tradition, Rome was at war 
with the Latin towns, and carried it on successfully. How many 
of the particular events which are recorded may be historical, 
I am unable to say ; but that there was a war is credible enough. 
Ancus, it is said, carried away after this war many thousands 
of Latins, and gave them settlements on the Aventine. The 
ancients express various opinions about him ; sometimes he is 
described as a captator auroe pojmlai-is ; sometimes he is called 
bonus Ancus. Like the first three kings, he is said to have 
been a legislator, a fact which is not mentioned in reference to 
* Namely 4 + 5, five being the plebeian number. 


the later kings. He is moreover stated to have established the 
colony of Ostia, and thus his kingdom must have extended as 
far as the mouth of the Tiber. 

Ancus and Tullus seem to me to be historical personages; 
but we can scarcely suppose that the latter was succeeded by 
the former, and that the events assigned to their reigns actually 
occurred in them. These events must be conceived in the fol- 
lowing manner. Towards the end of the fourth reign, when, 
after a feud which lasted many years, the Romans came to an 
understanding with the Latins about the renewal of the long 
neglected alliance, Rome gave up its claims to the supremacy 
which it could not maintain, and indemnified itself by extend- ' 
ins its dominion in another and safer direction. The eastern 
colonies joined the Latin towns which still existed: this is 
evident, though it is nowhere expressly mentioned ; and a por- 
tion of the Latin country was ceded to Rome; with which the 
rest of the Latins formed a connection of friendship, perhaps 
of isopolity. Rome here acted as wisely as England did when 
she recognised the independence of Xorth America. 

Li this manner Rome obtained a territory. The many 
thousand settlers whom Ancus is said to have led to the Aven- 
tine, were the population of the Latin towns which became 
subject to Rome, and they were far more numerous than the 
two ancient tribes, even after the latter had been increased by 
their union with the third tribe. In these country districts lay 
the power of Rome, and from them she raised the armies with 
which she carried on her wars. It would have been natural 
to admit this population as a fourth tribe, but such a measure 
was not agreeable to the Romans : the constitution of the state 
was completed and was looked upon as a sacred trust, in which 
no change ought to be introduced. It was with the Greeks and 
Romans as it was with our own ancestors, whose separate tribes 
clung to their hereditary laws and diilc'red from one another in 
this respect as much as they did from the Gauls in the colour 
of their eyes and hair. They knew well enough that it was in 
their power to alter the laws, but they considered them as 
somethins which ought not to be altered. Thus when the 
emperor Otho was doubtful on a point of the law of inheritance, 
he caused the case to be decided by an ordeal or judgment of 
' God. In Sicily one city had Chalcidian, another Doric laws, 
although their populations, as well as their dialects, wore 


greatly mixed; but the leaders of tliose colonies had been 
Chalcidians in the one case, and Dorians in the others. The 
Chalcidians moreover were divided into four, the Dorians into 
three tribes, and their differences in these respects were mani- 
fested even in their weights and measures.^ The division into 
three tribes was a genuine Latin institution; and there are 
reasons which render it probable that the Sabines had a divi- 
sion of their states into four tribes. The transportation of the 
Latins to Rome must be regarded as the origin of the plebs. 


Although the statement that Ancus carried the Latins away 
from their habitations and transplanted them to Rome, as if he 
had destroyed their towns, cannot be believed because it is 
impossible, since the settlers would have been removed many 
miles from their possessions and would have left an empty 
country, yet it cannot be doubted that Ancus Martins is justly 
called the founder of the town on the Aventine. There arose 
on that spot a town which even to the latest times remained 
politically separated : it existed by the side of Rome but was 
distinct from it, not being included within the pomoerium so 
long as any value was attached to that line of demarcation. 
In following the narrative as it has been transmitted to us, 


we now come to a period, which was probably separated by a 
great chasm from the preceding one. In the reign of Tarquinius 
Priscus, Rome appears in so different a light, that it is impos- 
sible to conceive him as the successor of Ancus, whose conquests 
were confined to a small space, and under whom Rome formed 
its first connection with the sea through the foundation of 
Ostia; whereas under Tarquinius, things are mentioned of 
which traces are visible to this day. Tarquinius is described 
as half an Etruscan, the son of Damaratus by an Etruscan 
woman. His father is said to have been a Bacchiad, who in 

* When the Achacans spread over the Peloponnesus, Sicyon first adopted 
their vS/itfia, and its example was gradually followed by the other towns, and 
thus the Doric laws almost disappeared. Attempts were made to compel Sparta 
also to abandon its old laws, but without success. — Ed. 


the revolution of Cypselus quitted Corinth with his immense 
wealth, and went to Tarquinii. His property descended to his 
son L. Tarquinius, as his elder son Aruns had died previously, 
leaving a wife in a state of pregnancy, a circumstance of which 
the elder Tarquinius was not aware. This account is com- 
monly believed to be of considerable authority, because Poly- 
bius, though a Greek, mentions Tarquinius as a son of" 
Damaratus, and because chronology is supposed not to be 
against it. But this is only an illusion, because the time 
depends upon the correctness of the chronological statements 
respecting the Eoman kings, according to which Tarquinius 
Priscus is said to have ascended the throne in the year lo2 
after the building of the city; but if we find ourselves com- 
pelled to place him at a later period, the story of Damaratus 
and Cypselus, which may with tolerable certainty be referred 
to the thirtieth Olympiad, must fall to the ground. I have 
already remarked elsewhere, that the ancient annalists, with 
the sole exception of Piso, never doubted that Tarquinius 
Superbus was a son of Tarqiunius Priscus, whence the time 
assigned to the latter must be utterly wrong; his relationship 
to Damaratus is therefore impossible. 

The story of Damaratus belongs to the ancient tradition 
respecting the • connection between Greece and Etruria, and 
the civilisation introduced fuom the former into the latter 
country. ^V^lat Evander was to the Latins, that Damaratus 
was to the Etruscans or Tyrrhenians, as he is said to have 
made them acquainted with the Cadmcan alphabet ; and accord- 
ing to the most ancient Greek tradition he belongs to a period 
as remote as that of Evander. Wliat caused him to be connect- 
ed with Tarquinius Priscus was the fact of the ancient legend 
mentioning Tarquinii as the place where Damaratus settled, 
though it undoubtedly knew nothing of his belonging to the 
family of the Bacchiadac, which must be an addition made by 
later narrators, who every where endeavoured to connect the 
history of one country with that of another. The reason for 
making Damaratus proceed to Tarquinii may have been the 
fact that Tarquinii was an important city; but at the same 
time there is no doubt that a connection existed between Tar- 
quinii and Corinth. It was formerly believed that the vases 
and other vessels found in Tuscany were of" ICtruscan origin ; 
this idea was afterwards justly given up. but then a belief arose 

VOL. I. E 


that such vases never existed in ancient Etruria. In our days 
vessels are dug out of the ground at Corneto, which perfectly 
resemble the most ancient Greek ones; I do not mean those 
which were formerly called Etruscan, but those actually found 
in Greece and belonging to the earliest times, especially the 
Corinthian ones, of which representations are given in Dod- 
well.'' Pieces of such vases are found only in the neighbour- 
hood of ancient Tarquinii; in all the rest of Tuscany scarcely 
one or two of them have been discovered; in 'the north- 
eastern part of the country, about Arezzo and Fiesole, the 
Arretinian vases of red clay with raised figures are of quite a 
peculiar form and very numerous, but do not occur any where 
on the coast. This artistic connection between Tarquinii and 
Greece, especially Corinth, is accounted for in the tradition by 
the statement that the artists Eucheir and Eugrammus ac- 
companied Damaratus from Corinth. 

Now when it was observed that Tarquinius Priscus was 
referred to Tarquinii, and a comparison of this statement was 
made with the tradition that the solemn Greek worship had 
first been introduced by him into Eome, people at once said : 
This must be the work of an ancient Greek; they compared 
the Roman chronology, as it was laid down in the work of the 
pontiffs, with the chronology of Greece, a comparison which 
might be made after the time when Timaeus wrote his history. 
They soon found that the combination became possible, if 
Damaratus was represented as the father of Tarquinius. This 
Tarquinius Priscus or Lucumo is said to have gone to Eome 
with his wife Tanaquil, an Etruscan prophetess, because at 
Tarquinii he did not enjoy the full rights of a citizen. On 
his journey thither a marvellous occurrence announced to him 
that heaven had destined him for great things ; many glorious 
exploits are ascribed to his reign; but our narratives here 
diverge : that of Livy is very modest, but another represents 
him as the conqueror of all the Etruscan towns. All this 
may be read in detail in Dionysius, and the accounts of it 
belong to the ancient Roman annals, so that Augustus caused 
these victories to be registered even in the triumphal fasti as 
three distinct triumphs and with certain dates, as we see from 
the fragments.2 The Romans had the more reason to believe 

' Classical Tour, ii. p. 195. — Ed, 

' The destruction of this monument is the fault of those who made it; they 
ought to have chosen a better material. — N. 


these statements, because Tarqulnius Priscus was mentioned as 
the king who united the town of the Sabines to that of the 
Eomans and executed the gigantic works by which the valleys 
also were filled up. 

The same tradition invariably calls Tarquinius Priscus, 
Lucumo; now this never was a proper name, but was the 
Etruscan title of a prince. Whenever the Romans wished to 
invent a story about the Etruscans, they called the men Lucumo, 
Aruns, or Lars. The last probably signifies " king." Aruns 
was an ordinary name, as we know from the inscriptions on 
Etruscan tombs, in which we can distinguish the names though 
we do not understand a single word. I have examined all 
the Etruscan inscriptions; and have come to the conclusion 
that their language is totally different from Latin, and that 
only a few things can be made out by conjecture, as for 
example, that ril avil means vixit annos. Lucumo does not 
occur in these inscriptions, and the ancient philologers, such 
as Verrius Flaccus, knew that it was not a name. The Romans 
had several traditions about a Lucumo, who is connected with 
the history of Rome; one, for example, was a companion of 
Romulus. All these Lucuraos are no other than Liicius Tar- 
quinius Priscus himself, that is, tradition has referred to him 
all that was related of the others. Livy says that at Rome he 
assumed the name Lucius Tarquinius Priscus, a statement for 
which scholars have charged Livy with rashness : it is rashness, 
however, only on the supposition that he took Priscus in the 
sense of "ancient." But it may often have happened to 
Livy when writing his first book, that he composed his narra- 
tive with the conviction that it was not all literally true, and 
that something else might be imderstood by it. Priscus was 
a common name among the Romans; it occurs in the family 
of the Servilii and many others; Cato was called Priscus before 
he obtained the name Cato, that is, Catus, the prudent. 
I am satisfied that Tarquinius was connected with the town of 
Tarquinii only on account of his name, and that he was in 
reality a Latin. This opinion is supported by the mention of 
Tarquinii, who after the expulsion of the kings dwelt at Lau- 
rentum, and by the statement, that Collatinus retired to 
Lavinium which was a Latin town. Moreover, the whole 
story of the descent of Tarquinius Priscus from Damaratus is 
overturned by the fact that Cicero, Varro, and even Livy, 

E 2 


acknowledged the existence of a gens Tarquinia : and how 
totally different is a gens from a family consisting of only two 
branches, that of the kings and that of CoUatinus? Yarro 
expressly says: Omnes Tarquinios ejecerunt ne quam reditionis 
per gentilitatem spem haberet. 

The desire also of accounting for an Etruscan influence 
upon Rome, contributed, independently of his name, towards 
connecting Tarquinius with Etruria. The Romans described 
Servius TuUius, who was an Etruscan, as a Latin of Corni- 
culum, and made Lucius Tarquinius Priscus, who was a Latin, 
an Etruscan. Thus the whole account of his descent is a fable, 
and Tanaquil too is a perfect fiction, for the Romans gave this 
name to any woman whom they wished to characterise as 
Etruscan, it being a common name in Etruria, as we see in 
the inscriptions. In the ancient native tradition, Tarquinius 
was married to a Latin woman, Caja Caecilia, a name which must 
be traced to Caeculus, the founder of Praeneste. Her image 
was set up in the temple of Semo Sancus, for she was worshipped 
as a goddess presiding over female domesticity. This has a 
genuine national character ; so much so that in the ancient 
legend the people are said to have rubbed off particles from 
the girdle of her brass statue, to be used as a medicine. 

It is therefore historically certain that there was a Latin of 
the name of Tarquinius Priscus ; but he most probably belonged 
to the Luceres, for whom he procured seats in the senate, one 
hundred being added, as gentes minores, to the two hundred 
senators who were called up by the king before the gentes 
minores ? In the insurrection of Tarquinius against Servius 
Tullius, these additional senators were his faction. The reign 
of Tarquinius, as I have already remarked, is probably sepa- 
rated by a great chasm ^from the preceding period, for under 
him Rome presents quite a different appearance from what it 
had been before. The conquests ascribed to Ancus ]\Iarcius 
are confined to a very small extent of country : he made him- 
self master of the mouth of the Tiber and fortified Ostia. 
But after him a state of things is described by the historians, of 
which traces are still visible. Even at the present day there 
stands unchanged the great sewer, the cloaca maxima, the 
object of which, it may be observed, was not merely to carry 
away the refuse of the city, but chiefly to drain the large^lake, 
which was formed by the Tiber between the Capitoline, 


Aventine and Palatine, then extended between the Palatine 
and Capitoline, and reached as a swamp as far as the district 
between the Quirinal and Viminal. This work, consisting of 
three semicircles of immense square blocks, which, though 
without mortar, have not to this day moved a knife's breadth 
from one another, drew the water from the surface, conducted 
it into the Tiber, and thus changed the lake into solid ground ; 
but as the Tiber itself had a marshy bank, a large wall was 
built as an embankment, the greater part of which still exists. 
This structure, equalling the pyramids in extent and massive- 
ness, far surpasses them in the difficulty of its execution. It 
is so gigantic, that the more one examines it the more incon- 
ceivable it becomes how even a large and powerfril state could 
have executed it. In comparison with it, the aqueducts of 
the emperors cannot be considered grand, for they were built 
of bricks with cement in the inner parts, but in the more 
ancient work everything is made of square blocks of hewn 
Alban stone, and the foundations are immensely deep. 

Now whether the cloaca maxima was actually executed by 
Tarquinius Priscus or by his son Superbus, is a question about 
which the ancients themselves are not agreed, and respecting 
which true historical criticism cannot presume to decide. But 
thus much may be said, that the structure must have been 
completed before the city encompassed the space of the seven 
hills, and formed a compact whole. This, however, was effected 
by the last king, and accordingly if we wish to make use of 
a personification, we may say that the great sewer was built in 
the time of Tarquinius Priscus. But such a work cannot 
possibly have been executed by the powers of a state such as 
Kome is said to have been in those times; for its territory ex- 
tended from the river not more than about ten miles in breadth, 
and, at the utmost, from thirty to forty in length, which is 
not as large as the territory of Niirnberg: and this objection is 
the more weighty when we take into consideration all the 
difficulties of an age in which commerce and wealth had no 
existence. A period must therefore have been passed over in 
our histories, and we now all at once see Rome as a kingdom 
ruling far and wide, and quite different from what it had been 
before. Of this extensive dominion no mention is made by 
Livy, though he expresses his wonder at these architectural 
structures ; he conceives that time still as the state of the city's 


infancy, and is therefore under the same erroneous belief as 
Cicero and all the later writers, namely, tliat the kingly 
period must be regarded as an age in which Kome was ex- 
tremely weak. The statement of Dionysius, that the Etruscan 
towns, the Latins, and the Sabinespaid homage to Tarquinius, 
might therefore seem to be more deserving of credit, but all 
the accounts of the manner in which this state of things was 
brought about, are so fabulous and fictitious, that it is evident 
they must have been manufactured by those who attempted 
to solve the mysterious problem; and we have no historical 
ground to stand upon. But in whatever way Tarquinius 
Priscus may have been connected with the Tuscan traditions 
about the conquests of Tarchon, we can with certainty say, 
that at that time Home was either herself the mistress of a 
large empire, or was the seat of a foreign ruler; at any rate, 
Eome at one period was the centre of a great empire. 

Another no less mysterious undertaking is ascribed to the 
reign of the same Tarquinius Priscus: he wanted, it is said, 
to double the Komulian tribes, that is, to add three new tribes 
with names derived from his own and those of some of his 
friends. This plan was opposed by the augur Attius Navius, 
because the three tribes were inseparably connected with the 
auspices. The tradition probably did not run as Livy relates it, 
but as we read it in Dionysius, that Tarquinius himself cut 
through the wet stone, and in doing so injured his hand. The 
king now is said, not indeed to have formed three new tribes, but 
to have added new centuries to the ancient ones.^ In this tradi- 

* No ancient nation could change its division without altering its whole charac- 
ter. This is not the case in modern states, forwhen at Florence the small guilds 
were added to the seven great ones, it produced little change, and even if the 
number had been changed to twelve, it would have been of no consequence. 
But if in antiquity an Ionic nation, for example, which had four tribes, had 
assumed a different dinsion, it would have been equivalent to a revolution, and 
coiild have been done only by entirely changing the character of the people. 
When the number of tribes at Rome had been reduced to twenty, and was after- 
wards raised to upwiu'ds of thirty, tliis was done because the inviolability of 
form had been given up in consequence of other circumstances. Cleisthenes is 
said to have increased the four Attic tribes to ten ; but I believe .that he never 
di-vidcd the demos into ten tribes, and that the throwing together of the deinos 
and the politae, whicli caused the four ancient tribes to disappear, took place at 
a later time. It is a singular fact, that we can describe the ancient Roman con- 
stitution witli much more certainty than that of Athens at the same period, 
although tlie extant Attic historians lived scai-cely a century after the great 
change*. -N. 


tion, therefore, mention is made of the unalterableness of the 
tribes, and of the ruler's intention to double the population by 
the admission of new citizens, an intention which the ancient 
citizens opposed by an appeal to the sacred rites of religion. 
But we here see a ruler who is not a mere magistrate, but one 
who has it in his power to give weight to his authority: as 
far as the form was concerned he yielded, but he virtually 
introduced the change by forming second centuries. Centuries 
and tribes were originally identical, each tribe containing one 
hundred gentes ; but what these second centuries were, is 
quite uncertain. One supposition is, that, as many of the old 
gentes had become extinct, Tarquinius formed new ones ; 
supposmg, for example, that those of the Ramnes had thus 
been reduced to fifty, the king would have added fifty new 
gentes as secundi Ramnes, to complete the number one hun- 
dred. We have an example in the Potitii,who became extinct 
in the time of Appius Claudius, when, it is said, they still 
consisted of twelve families. The history of exclusive families 
shows how rapidly they become extinct : in Styria there were 
formerly 2,000 noble families; at present there scarcely exist 
a dozen : in the duchy of Bremen the nobles entitled to take 
part in the diet were within fifty years reduced to one half of 
their original number, only because they tolerated no marriages 
except among themselves. In Liineburg the government was 
formerly in the hands of the houses {gentes), of whom at 
present only one remains. It is not impossible that Tar- 
quinius may have united the remnants of the ancient curiae, 
and then supplied the number of the wanting gentes. What 
recommends this conjecture, is the fact that there continued 
to be some difference between the old and the new gentes ; the 
new centuries certainly had not so much influence as they would 
have had, if they had been constituted as separate tribes. 

It is a very dangerous thing to seek for allegories in histo- 
rical statements, and then to presume to derive from them 
historical facts. Thus as Ancus Marcius was the creator of 
the plebs, and as Tarquinius is said to have been miu'dered by 
the JMarcii, we might infer that Tarquinius, who belonged to 
the Luceres, and had introduced them into the senate, perished 
in an insurrection of the plebeians. But this is a most hazard- 
ous conjecture, and for this reason I have not printed it in my 
history. In mentioning it here I rely on that confidence 


wliich a man may claim who has devoted himself to these inves- 
tigations for eighteen years almost uninterruptedly, and who 
even before that time had with fondness spent many a year 
upon them. Do not mistake possibilities for historical results. 
The tradition which represents Tarquinius as the acknow- 
ledged head of the twelve Etruscan towns leads us to speak 
about the Etruscans. Of all the nations of antiquity they are 
perhaps the one concerning which the most different things 
have been said, though our materials are of the slenderest kind, 
and concerning which accordingly the greatest misconceptions 
have been formed. The impositions of such persons as Annius 
of Viterbo, Inghirami and others, are of the most impudent 
character, and yet have become the groundwork of many later 
productions : they misled Dempster, and through him Winckel- 
mann was deceived. In the eighteenth century the ItaHans 
ceased indeed to forge documents, but with the greatest con- 
ceit they pretended to explain the inexplicable. Many Etrus- 
can monuments with inscriptions exist, but few are large. 
Five years ago an altar was dug out which is covered on 
three sides with inscriptions; a cippus was found at Perugia, a 
coffin at Bolscna, etc. These monuments have been published 
either separately or in collections, particularly by Lanzi; some 
works of art also bear inscriptions; to interpret them has a 
great charm, because if we could read them, a new light would 
be thrown upon our investigations. This has given rise to 
the confident assertions that they can be explained, and the 
most arbitary interpretations have been put upon them. The 
Eastern languages and the Celtic have been resorted to for 
assistance, until at length Lanzi proceeded on the supposition 
that the Etruscan was a kind of Greek, and, contrary to all 
the rules of grammar, arbitrarily made out some bad Greek; 
with all our Etruscan monuments, we know nothing, and are 
as ignorant as we were of the hieroglyphics previously to the 
time of Champollion ; nothing but large bilingual inscriptions 
can be of any assistance. We may say with certainty that the 
Etruscan has not the slightest resemblance to Latin or Greek, 
nay, not to any one of the languages known to us, as was 
justly remarked even by Dionysius. This passage of Diony- 
sius has been intentionally overlooked, or its positive meaning 
has been distorted into a conditional one. The Umbrian on 
the Eugubinian tables rcscinblcs Latin. 


Dionysius states that the Etruscans looked upon themselves 
us an original people descended from no other race, and which 
called itself Easena* and knew nothing of the names Tyrrhen- 
ians and Etruscans ; nor of the Grecian traditions respecting 
them. But the Greeks had two distinct traditions about the 
Tyrrhenians, which they referred to the Etruscans : the one 
recorded by Hellanicus stated that Pelasgians from Thessaly 
had settled at Spina at the mouth of the Po, whence they 
proceeded across the mountains into Etruria; according to the 
second related by Herodotus, the Lydians, in the time of 
Atys, are said to have been visited by a famine, whereby a 
part of the people was obliged under Tyrrhenus to emigrate 
to Italy. This latter statement is controvei'ted by Dionysius 
with that sound criticism which we sometimes meet with in 
his work, that neither the laniTuacre nor the religion of the 
Etruscans bore any resemblance to that of the Lydians, and 
that neither the Etruscans nor the Lydian historian Xanthus 
were acquainted with it.^ Dionysius here saw con-ectly, 
because he was not confined to books, but could judge from 
personal observation. The other tradition he treats differently ; 
he does not give it up, but refers it to the Aborigines and not 
to the Etruscans. The Italian antiquaries, on the other hand, 
have either clung to the Lydian tradition, or referred the 
emigration of the Pelasgians from Thessaly to the Etruscans, 
and they say that the inhabitants of Cortona (Croton) were 
not at all different from the neighbouring tribes, notwithstand- 
ing the protestation of Herodotus. I can here give only the 
results of my investigations about the Etruscans. In the new 
edition of the first volume of my Roman history, I have 
proved that the name Tyrrhenians was transferred by the 
Greeks to the Etruscans, just as we use the name Britons 
when we speak of the English, or INIexicans and Peruvians in 
speaking of the Spaniards, in America, because the Britons, 
Mexicans and Peruvians originally inhabited those countries, 
although a new immi2:ratinoj nation has established an order 
of things so entirely new that we perceive no more traces of 

* Easena, probably not Rasenna; Has is the root and ena the termination: as 
in Porscna, Caecina; the Etruscans, like the Semitic nations, did not double the 
consonants. — N. 

* C. O. Miiller has shewn that the work of Xanthns was undesen-cdlv looked 
upon by the Greeks as spmious.^N. 


an earlier condition than if it had not existed at all. The 
Tyrrhenians were a people quite different from the Etruscans, 
but inhabited the sea coast of Etruria, as well as the whole 
southern coast, as far as Oenotria, that is Calabria and Basili- 
cata. These Tyrrhenians were Pelasgians just as much as 
those of Peloponnesus and Thessaly, and when we read in 
Sophocles of Tvpprjvoi UeX.aa'yol in Argos, when according 
to Aeschylus, Pelasgus, son of Palaechthon, ruled in Argos, 
when according to Thucydides Tyrrhenians lived on mount 
Athos and in Lemnos, and according to Herodotus at the foot 
of Hymettus, we must recognise them everywhere as branches 
of the same stock. In the history of Asia Minor there is a 
gap beginning after the destruction of Troy, and we must 
fill it up by supposing that the Lydians, Carians and Mysians 
advanced from the interior towards the coast into the territory 
of Troy, and that the Maeonians and other Pelasgian tribes 
were partly subdued and partly expelled. The Maeonians, 
who are always distinguished from the Lydians, were likewise 
Tyrrhenians, and are so called by Ovid in the fable of Bacchus. 
These were the Tyrrhenians that gave their name to the 
western coast of Italy and to the Tyrrhenian sea, and whom 
the Komans called Tusci. Both names were afterwards trans- 
ferred to the Rasena who descended as conquerors from the 
Alps. This view at once renders the account of Herodotus 
perfectly clear, and is now generally adopted both in Germany 
and England. The tradition in Herodotus is a genealogy 
intended to explain how it happened that Lydians existed 
in Italy as well as in Lydia. 

There is one difficulty, which though it does not weaken 
the evidence of my view, is nevertheless a surprising fact, 
namely, that after the Etruscan conquest of the Tyrrhenian 
country, the language of the Rasena is the only one that is 
found on the many monuments, and that we do not find a 
trace of inscriptions in a language akin to the Greek, such as 
we must suppose the Tyrrhenian language to have been. But 
in the first place, almost all the inscriptions have been dis- 
covered in the interior of the country, about Perugia, Volterra, 
Arrczzo and other places, where the original population was 
Umbrian and only a very few on the coast about Pisa, Popu- 
lonia, Caere and Tarquinii ; some more have lately been found 
at Tarquinii, but have not yet been published. We might 


also say, although no Tyrrhenian inscriptions have hitherto 
been found, still they may yet be discovered; but such an eva- 
sion is worth nothing. Under the rule of a conquering nation 
which imposes a heavy yoke on the conquered, the language 
of the latter frequently becomes quite extinct : in Asia and 
many other countries, it was the practice to forbid the use of 
the vernacular tongue, in order to prevent treachery. The 
Moors were, in many respects, mild rulers in Spain, and the 
country flourished under tjiem ; but in Andalusia one of their 
kings forbade the Christians to use the Latin language^ under 
penalty of death, the consequence of which was that a hun- 
dred years later not a trace of it occurs. The whole Christian 
population of Caesarea spoke Greek down to the eighteenth 
century; when a pasha prohibited it, and after the lapse of 
thirty or forty years, when my father visited the place, not one 
of the inhabitants understood Greek. When the Normans 
conquered Sicily, the only languages spoken in the island were 
Greek and Arabic, and the laws were written in Greek as late 
as the time of the emperor Frederic II., but afterwards it dis- 
appears all at once. The same thing happened in Terra di 
Lecce and Terra di Otranto, where afterwards the names were 
Italian, while the language of common life remained Greek, 
until 200 years later, in the fifteenth century, it died away. 
In Pomerania and Mecklenburg the Wendic language dis- 
appeared within a few generations, and that without an immi- 
gration of Germans, but merely because the princes were 
partial to the German language; the conquerors of Branden- 
burg forbade the use of Wendic under penalty of death, and 
in a short time nothing was spoken but low German. The 
Etruscans had quite an aristocratic constitution, and lived in 
the midst of a large subject country; under such circumstances 
it must have been of great importance to them to make their 
subjects adopt the Etruscan language. 

The conquering Easena must have come down from the 
Alps, since according to Livy and Strabo the Kaetians as well 
as the other Alpine tribes, the Camuni, the Lepontii on the 
lake of Como, and others, belonged to the race of the Etrus- 
cans. No ancient writer has ever asserted that they withdrew 
from the plains into the Alps in consequence of the conquests 
of the Gauls, and it would be absurd to think that a people 


which fled before the Gauls from the plain of Patavium, should 
have been capable of subduing Alpine tribes, or should have 
been tolerated among them, unless the Alpine districts had 
before been in the possession of their kinsmen. There is a 
tradition, probably derived from Cato, that the Etruscans 
conquered 300 Umbrian towns ; these towns must be conceived 
to have been in the interior of Tuscany, a part of which bore 
the name of Umbria for a long time after ; and a river Umbro 
also is mentioned. The Etruscans therefore are one of the 
northern tribes that were pushed southward by the pressure of 
those early migrations of nations which are as well established 
in history as the later ones, although we have no written re- 
cords of them; they were migrations like that which had 
pressed forward the lUyrianSj in consequence of which the 
Illyrian Enchelians about the fortieth Olymiad penetrated 
into Greece and plundered Delphi, as Herodotus relates. 
Such a migration must have driven the Etruscans from the 
north. They at one time inhabited Switzerland and the 
Tyrol; nay, there can be no doubt that the Etruscans in those 
countries experienced the same fate as the Celts in Spain, and 
that some tribes maintained themselves there longer than others. 
The Heidenmauer (the heathen-wall) near Ottilienberg in 
Alsace, which Schweighiiuser has described as one of the most 
remarkable and inexj)licable monuments, is evidently an Etrus- 
can work ; it has exactly the character of the Etruscan fortifi- 
cations, such as we find them at VolteiTa, Cortona, and Fiesole. 
Some have called this kind of architecture Gallic, but without 
any foundation, as we sec from Caesar's description, as well as 
from other ruins and buildings in Gaul. 

In central Italy there are two essentially different modes of 
fortification; the one consists of what are commonly called 
Cyclopean walls formed of polygonal stones which are put 
together intentionally without any regular order; such a wall 
is raised around a hill so as to render it abnost perpendicular, 
but on the top of the hill there is no wall; a path (clivus) 
accessible on horseback leads to the top and there are gates botli 
below and above. In this manner the Koman and Latin hills 
were fortified. The other kind of fortification is Etruscan: 
on the highest ridge of a hill difficult of access, a wall is built, 
not of polygons but of parallclopipcda of extraordinary dimen- 
sions and very rarely of ^;cj[uarc blocks; the wall runs along 


the ridge of the hill in all directions; such is the case at 
Volterra, and of the same kind is the above mentioned wall in 
Alsace. I do not place the construction of the latter in a 
very remote time, but conceive it to be the work of a tribe 
akin to the Etruscans, which long maintained itself in that 
country against the Celts; although I must add, that I should 
not like to refer to the wall in question as an irrefragable 
argument for the existence of such a tribe in that district. 
Now the Etruscans first settled in twelve towns in Lombardy, 
extending to about the present Austrian fi-ontier towards Pied- 
mont (Pa via was not Etruscan), in the south from Parma to 
Bologna, and in the north from the Po to Verona; they then 
spread farther, and in the country south of the Apennines they 
either founded or enlarged twelve other towns, from which 
they ruled over the country. The common opinion is, that 
the Etruscans were a very ancient people in Italy, and I 
myself entertained tliis ^aew for a longtime; but in Tuscany 
they were not very ancient, and in the southern part of Tus- 
cany, which now belongs to the papal dominion, they did not 
establish themselves till a very late period. Herodotus relates, 
that about the year of Kome 220 the unfortunate Phocaeans 
were conquered in a sea fight by the Agyllaeans of Corsica 
and the Carchedonians, and that those of them who were 
taken prisoners were stoned to death. When Heaven punish- 
ed the Agyllaeans for this cruelty, they sent to Delphi, and 
Apollo ordered them to offer Greek sacrifices, and worship Greek 
heroes. Now all writers are unanimous in stating that Agylla 
bore this name as long as it was Pelasgian, and that afterwards 
the Etruscans called it Caere. We may with great probability 
look upon Mezentius, the tyrant of Caere in^the legend which 
Virgil ^vith his great learning introduced into his poem as the 
Etruscan conqueror of Caere; he afterwards appears as the 
conqueror of Latium, and demands for himself the tenth of its 
wine or even the whole produce of the vineyards. The Etrus- 
can conquests belong to the period of the last kings of Rome, 
and are connected with the expeditions of the Etruscans against 
Cuma and into the country of the Volscians; they spread 
into these districts about the time between the sixtieth and 
seventieth Olympiads; according to Cato's statement, which is 
certainly of great weight, they founded Capua in the year of 
Rome 283. The shortness of the time in which that town is said 


to have risen to greatness and declined again, which Velleius 
mentions as an objection, cannot render the fact improbable. 
Capua had after all existed for 250 'years before it became 
great, and New York is a far more surprising instance of rapid 
growth. The flourishing period of this peofile therefore was the 
time when Hiero of Syracuse defeated them near Cuma, and 
they began to decline at the beginning of the fourth century 
of Kome ; the Eomans were then rising, and about the middle 
of that century the Gauls deprived the Etruscans of the northern 
part of their dominion, their possessions about the river Po. 

AVlien people began to perceive that the Alban origin of 
Rome could not be maintained, Rome was looked upon as an 
Etruscan colony, and I myself brought forward this opinion. 
It forms the ground-work of the first edition of my history, 
because I then considered the Albano-Latin origin to be 
erroneous; the Etruscan origin seemed to me to be confirmed 
by several circumstances, particularly by the statement of one 
Volnius in Varro, thatthe names of the earliest Roman tribes 
were Etruscan; and also by the observation that the secret 
theology of the Romans had come from Etruria, that the sons 
of the first ten in the Roman senate learned the religious laws 
in Etruria ; and lastly, that the worship of Jupiter, Juno,^and 
Minerva in the Capitol was probably of Etruscan origin. But 
an unbiassed examination^afterwards convinced me that this 
theory was unfounded; and that the two original elements of 
the Roman state were Latin and Sabine (though I do not wish 
to dispute the later addition of an Etruscan element), that 
Rome is much older than the extension of the Etruscans in 
those districts, and consequently that either the statement of 
Volnius is groundless, or the names of the tribes are of a more 
recent date than the tribes themselves, and lastly, the great influ- 
ence of the Etruscans about the time which is commonly desio-- 
natsd as the reign of Tarquinius Priscus and Servius Tullius is 
perfectly sufficient to explain all the Etruscan institutions at 
Rome. No ancient writer ever speaks of an Etruscan colony 
at Rome. The only question now is, whether the Etruscans 
extended their dominion at so early a period, that even in the 
time of Tarquinius Priscus they were in possession of Tarquinii 
and the neighbouring places, or wliether they did not begin to 
appear about and beyond the Tiber till the sixtieth Olympiad. 

Before we proceed to describe the changes which took place in 


those times, we must give the history of the Etruscans as far as it 
is known, and add a sketch of the earliest constitution of Rome. 
All we know of the history of Cuma is very obscure. Its 
foundation is assigned to a period more remote than that of 
any other Greek city in that district, which could not have 
been done had not Cuma ceased at an early period to be a 
Greek town and come into the hands of the Oscans, before 
people in that country began to write Greek. All towns 
undoubtedly had eras from their foundation, the fixed chrono- 
logical data furnished by which were afterwards reduced to 
Olympiads; for it was not till very late that the Greeks began 
to reckon according to Olympiads. Timaeus (Olymp. 120 — 
130) was the first who did so; Theophrastus did not. Now 
where a town like Cuma was lost to the Greeks, they had no 
trace of the era of its foundation, nor anything to take as a 
guide except the genealogies of its ctistae (/cTio-Tat.) When 
therefore, it was stated that this or that person had founded a 
town, they ascended genealogically backward to Troy and the 
heroes; and this is the reason why Cuma was thought to be 
200 years older than the surrounding Greek towns: its era 
had been lost very early, but it was certainly not older than 
those of the other cities of similar origin. All that was known 
about Cuma probably existed in Neapolitan chronicles, of 
which Dionysivis made use. His description indeed of the 
war waged by the Etruscans against Cuma is mythical, for the 
Volturnus is said to have flowed back towards its source and 
the like; but this is a secondary matter; Herodotus too is 
mythical, when he describes the destruction of the Cartha- 
ginian army which fought against Gelon, but the occurrence 
of the war itself is not on that accoimt to be doubted. About 
the sixty-fourth Olympiad, the Cumans were in their highest 
prosperity and in possession of Campania; if therefore the 
Etruscans besieged Cuma at that time, they must then have 
been a conquering nation, a fact which beautifully agrees with 
Cato's statement, that Capua existed only 260 years after its 
foundation, meaning that it then became an Etruscan colony. 
We thus obtain the period from 250 to 280 years after the 
building of Rome (according to our common chronology) as 
the time during which the Etruscans must have crossed the 
Tiber. Between A.u. 220 and A.U. 230, Herodotus represents 
Agylla as a town which consulted the oracle of Delphi; but 


that the Etruscans, who were so proud of their own religion, 
should have done so, is wholly inconceivable, more especially 
as there existed an inveterate hatred between Etruscans and 
Greeks; hence the Romans received from the libri fatales 
which were of Etruscan origin, the command to sacrifice a 
Greek man and woman and a Gallic man and woman/* This 
national hatred shews itself everywhere, as in Pindar and in 
the Bacchic fable, where things are said of the Tyrrhenians 
which must be referred to the Etruscans. The Etruscans 
accordingly appear on the Tiber much later than is commonly 
supposed; they gradually extended their sway, attained the 
height of their power, maintained it for two generations, and 
then declined with ever-increasing speed. 

The early Etruscan history is scarcely known to us at all; 
in Tuscany we find twelve towns, perfectly independent of 
one another, yet at times united in common undertakings. 
Each of these towns was governed according to custom by a 
king, but there is no trace in any of the Italian nations of 
hereditary monarchies such as we see in Greece; these towns 
moreover formed no artificial confederacy, but a league some- 
times arose spontaneously, when they were assembled for 
common deliberation near the temple of Voltumna; they had 
also one priest, who presided over the whole nation. It seems 
probable — as the Romans did not understand the Etruscan 
language, we must take their statements with great caution — 
that in general enterprlzes one of the kings yyas chosen, whose 
sovereignty the other towns recognised, and to whom they 
gave np the ensigns of royalty ; but this distinction does not 
appear to have always been the result of an election, the 
supremacy being often assumed by some one town; thus 
Clusium was the capital of Etruria in the war with Porsena. 
Our historians conceive Rome to have stood in the same rela- 
tion to these Etruscan towns, which arc said to have sent 
to Tarquinius Prisons, or, according to others, to Servius 
Tullius, the ivory throne and the kingly insignia. Keither 
story is historically true, but it is an indication that under her 
last kings Rome was at the head of a mighty empire, which 
was much larger than in the first 160 years of the republic: 
and of which Rome itself still preserves traces. It seems 
to have been recognised as the capital more particularly in 
•^ Liv. xxii. 57. It was not from the Sibylline books, as Plutarch says. — N. 


relation to Etriiria, but this is only a transitory circumstance 
which may have been changed several times even under the 

The Etruscans bear all the marks of an immigrating people, 
and were probably not much more numerous than the Germans, 
who at the beginning of the middle ages settled in Italy. The 
towns possessed the sovereignty, and in the towns themselves, 
the burghers. The territories of the towns were large but had 
no influence; and it was this very oligarchical form of govern- 
ment, which rendered Etruria weak by the side of Rome, 
since arms could not be put into the hands of the people with- 
out danger. 

Dionysius, who very carefully gives us the exact expressions 
of his authorities, says, that the magnates of the Etruscans 
assembled with their clients for war. Among the Romans to 
enlist the clients, was only a last resource when the plebeians 
refused to go out to fight. Other circumstances also suggest 
that Etruria was inhabited by clients under a territorial aristo- 
cracy. During the advance of the Gauls, when the people on 
the left bank of the Tiber deserted Rome, she attached to 
herself those on the right bank ; Caere obtained the isopolity ; 
and four new tribes were formed of those who during the war 
had deserted Veii and Falerii." The historv of the insuri'ec- 
tion of Yulsinii also shows the people in the condition of 
subjects, as I have shewn in the first volimie of my Roman 
history. The Vulsinians gave to their clients the constitution 
of a plebs in order to Avard off the Romans; the plebs after- 
wards crushed their former masters, and the latter then threw 
themselves into the arms of the Romans, and allowed them to 
destroy their town. Such an oligarchy existed everywhere, 
whence we find so small a number of towns in Etruria, the 
whole country from the Apennines to Rome containing no 
more than twelve. The power of the nation therefore was 
only in the first stage of its development; there was no con- 
tinuous and growing life, nor any elements of a national 
existence as among the Romans and Samnites, who evidently 
did not oppress the old Oscan population, but became one 

'' They were evidently not formed of tranafngcie, as Lhy says, but of wliolc 
tribes which joined Rome in order to escape oppression; tliis is perfectly accord- 
ing to analogy, for only two tribes were formed out of the Volscians, and the 
same number also out of the Sabincs. — N. 
VOL. T. F 


with it and even adopted their language. The Lucanians, on 
the other hand, who were a branch of the Sahines, stood in 
quite a different relation to the ancient Oenotrians, for other- 
wise the number of their citizens would have been very different 
fi;oni that mentioned by Polybius. Opposite kinds of policy 
in thersC cases bear opposite fruits. The insurrection of the 
Bruttians was nothing else than a revolt in which the Oeno- 
trians, who had been clients even under the Greeks, broke 
their chains, when they came under new lords who treated 
them still more harshly. The Etruscans, notwithstanding 
their wealth and greatness, could not keep their ground against 
the Romans ; their towns did not form a closely united state 
like that of the Latins, nay, not even like that of the Achaeans, 
and in the fifth century of Eome, most of them laid down 
their arms after one or two battles; the only exception was 
that very Vulsinii where the clients had been changed into a 
plebs and which defended itself for thirty years. The Sara- 
nites resisted Rome for seventy years, but the Lucanians for 
only a very short time. 

The Etruscans have been treated with great favour by the 
moderns, but the ancients shewed them little respect. Among 
the Greeks, very unfavourable reports were current about their 
licentiousness and luxurious habits, although in regard to art, 
justice was done them to some extent; the perfection of all 
the mechanical parts of art and the old-fashioned forms had a 
great charm ; the signa Tuscanica were as much sought after 
in Rome as old pictures are now among ourselves. 

The Etruscans were esteemed especially as a priestly people, 
devoted to all the arts of prophecy, especially from meteoro- 
logical and sidereal phenomena, and from the entrails of victims : 
the art of discovering the future by augury was the peculiar 
inheritance of the Sabellian people. All this must surely be 
regarded as a wretched system of imposture. I will not deny 
that the observations of lightning led the Etruscans to interest- 
ing discoveries: they were aware of the lightnings which 
flash forth from the earth, and which are now acknowledged 
by naturalists, but were denied as late as thirty years ago. I 
am noAv less than formerly inclined to believe that they were 
acquainted with conductors of lightning; such knowledge 
would not have been lost so easily ; moreover it is not said that 
they attracted lightnings, but that thoy called them forth. 


In history, the Etruscans show themselves in anytliing but a 
favourable light; they were unwarlike and prone to withdraw 
from an impending danger by acts of humiliation, as in modern 
times so many states have done between the years 1796 and 
1813. The descriptions of their luxurious habits may be 
exaggerated, but they are not without foundation; for nearly 
two centuries they lived in the most profound peace under the 
dominion of Rome and exempt from military service, except 
in extraordinary emergencies, as in the Hannibalian war; and 
it must have been during that period that they possessed 
the unmense wealth and revelled in the luxuries of which 
Posidonius spoke. 

The Etruscans also had annals, of which the emperor Claudius 
made use; and some few statements may have been taken from 
them by Verrius Flaccus and Yarro. Their most celebrated 
hero is Caeles Vibenna, who is the only historical point, 
properly speaking, which we know in the history of the 
Etruscans. Caeles Vibenna is said by some to have come to 
Rome and to have settled on the Caelian hill ; but accordingr 
to others, who followed the Etruscan traditions, he died in 
Etruria, and his general, Mastarna, led the remnant of his 
army to Rome, where he is said to have named the Caelian 
hill after his own commander. Caeles always appears, in our 
accounts^ as a condottiere, as an independent general of a 
gathered host, unconnected with the towns, just like the Cata- 
lonian hosts at the beginning of the fourteenth century, and 
the East Indians in the eighteenth. His subsequent fortunes 
are not known ; but the emperor Claudius states, from Etrus- 
can books, that his faithful general, Mastarna, having gone to 
Rome and settled on Mount Caelius, was received into the 
Roman state under the name of Servius Tullius. This is 
possible enough, whereas the Roman tradition about Servius 
Tullius lies entirely within the sphere of the marvellous. The 
god of fire, it is said, appeared to Tanaquil in the ashes on the 
hearth, whereupon she ordered her maid to lock herself up 
there in bridal attire; the maid became pregnant and gave 
birth to Servius Tullius. As a sign of his descent from the 
god of fire, his head was surrounded by a fiery halo whenever, 
during his infancy, he was asleep; and in the conflagration of 
a temple his wooden statue which was within remained unin- 
jured. Conceited expositors have cautiously attempted to give 

F 2 


to tins narrative also tlie appearance of history : many avIio 
think his descent from a servant maid inconsistent, make him 
the son of a noble of Corniculum, who is said to have died, 
leaving behind him his wife in a state of pregnancy, in which 
she was taken to the king's palace. Others say that his mother 
indeed was a servant, but his father a king ; the fiery halo 
also is interpreted as a symbol of his precocious mind; non 
latuit scintilla ingenii in puero, as Cicero says. But the ancient 
poets were in earnest and did not mean any such thing. We 
have the choice; Ave may either leave the origin of Servius 
Tullius in obscurity, or believe that the Etruscan traditions are 
true. I am of opinion that Etruscan literature is so decidedly 
more ancient than that of the Romans, that I do not hesitate 
to give preference to the traditions of the former. As Tar- 
quinius Priscus was represented to be an Etruscan, merely 
because it was clear that there existed an Etruscan element at 
Rome, which on account of his name was referred to Tarquin- 
ius, so people described Servius Tullius as belonging to another 
race, especially as Rome would not be indebted to an Etruscan 
for the important reforms ascribed to this king. But as they 
could not connect him with any distinct gens, they went back 
to mythology and represented him like Romulus as the son of 
a god, and like Numa as the husband of a goddess. The 
mother is of no consequence to the son of a god.'' We cannot, 
however, draw any further inferences; for the statement that 
he was an Etruscan and led the remnant of Caeles Vibenna's 
army to Rome is of no historical value. Livy speaks of a 
war Avith Veii, but only in a hasty manner ; from which it is 
evident he kncAv it to be a mere forgery in the Fasti. 

In the tradition, Servius appears as a Latin who obtained 
possession of the throne not even by a regular election : to him 
are traced all the political laws, as all the religious laws are 
to Numa, a proof that neither of them appeared as an historical 
individual even to Livy. The gens Tullia, to Avhich Servius 
must have belonged either by birth or by adoption, is expressly 
mentioned as one of the Alban gentes that settled on the 

* The above passage respecting the Etruscan origin'of Servms Tiillius belongs 
to the lectures of the year 1826, but I was unwilhng to suppress it, although 
further on (p. 99) we have a ditferent view, taken from the lectures of the yeai- 
1828. The discussion licre introduced may be compared witli tliat in tlic Rom. 
Hist. vol. i. p. ."^S"), etc., but the above is clearer and more definite. — Ed. 


Caelius, and accordingly belonged to the Luceres; thus avc 
have here a king of the third tribe, or, since this tribe was 
closely connected with the commonalty, the throne is occupied 
by one of the commons who is said to have come from Corni- 
culum. He obtained the sovereignty without an election, but 
was afterwards recognised by the curiae. Historical facts may 
be embodied in this tradition ; but it is difficult to guess what 
legal relations were intended to be expressed by this story. 
Servius is important in three respects; he gave the city the 
legal extent which it retained down to the time of the emperors, 
though suburbs were added to it; he was the author of a 
constitution in which the plebs took its place as the second 
part of the nation; and he established an equal alliance with 
the Latins, who previously had been either in a state of war 
with, or of compulsory dependence on, Rome. 

In these respects, Servius is so important that we cannot 
help dwelling upon him. For the sake of greater clearness, 1 
shall here treat of Tarquinius Prisons and Servius TuUius as if 
they were historical personages, their names representing men 
who though not known to us, really existed, and in fact serving 
the same purpose as x, the symbol of an unknown magnitude 
in mathematics; we shall thus, as I have already remarked, 
start from the earliest appearance of Rome previously to the 
change ascribed to Servius. 

In its primitive form Rome was a town on the Palatine 
surrounded by a wall and ditch, with a suburb and a Sabine 
town on the Quirinal and the Tarpeian Hill. Rome grew out of 
the union of the two towns whose imited citizens were subse- 
quently designated by the common name of Romans. Servius 
combined into one whole that which before was divided into 
parts, and inclosed the city on all sides with fortifications and 
walls no less than live miles in circumference. The accounts 
of this wall and moat are not fables; the wall was perfectly 
preserved as late as the time of Augustus and Pliny, so that 
there was no room for fiction. Dionysius, who generally 
derived his materials from books, cannot have been deceived 
here, for he must have often seen the wall, it being a common 
promenade for the Romans. Rome then, in the time of Ser- 
vius, was a city as large as Athens after the Persian war, and 
in our days would be accounted a place of considerable im- 


All modern states, with tlie single exception of tlie canton 
of Schwyz, in tlieir governments and divisions have reference 
to territorial circumstances. Each town is divided into districts 
and wards; and in constitutional governments the representa- 
tion is based upon these divisions; whoever lives in a district 
elects and may be elected in it. But the ancients viewed the 
soil only as the substi'atum of the state, which they were of 
opinion existed in the individuals, so that certain associations 
gave a different character to the relation in which individuals 
stood to the state. Accordingly the state w^as divided into a 
number of associations, each of which again consisted of several 
families. Every one of these associations had its own assem- 
blies, courts, religious rights, laws of inheritance and of other 
matters. Whoever belonged to one transmitted these peculi- 
arities to his children and wherever he might live, whether 
within or without the state, he always belonged to that asso- 
ciation. But those who did not belong to it by birth, could 
be admitted only by a deviation from the rule, if the association 
permitted it. A person might be admitted into the state with 
all the rights which the ancients limited to the citizens as such, 
the rights, for instance, of acquiring landed property and of 
appearing in the courts of justice; and yet if he did not be- 
long to an association, he was only a pale-burgher, that is, he 
could not be invested with any office and was not allowed to 
vote. This Avas the principle of the earliest states of antiquity, 
the power of the state in this particular being limited to giving 
civil rights, or the rights of a pale-burgher, the state could 
not order an association to receive this or that individual as a 
member. In many states even the associations themselves 
had no power to admit a person, as, for example, where there 
existed close castes, among which there was no right of inter- 
marriage. Such an association, consisting of a number of 
families, from which a person may withdraw, but into which he 
either cannot be admitted at all or only by being adopted by the 
whole association, is a gens.^ It must not be confounded with 
OUT family, the members of which are descended from a com- 
mon ancestor; for the patronymic names of the gentes are 
nothing but symbols, and are derived from heroes. *° I assume 

^ The German word is ein Geschlecht. See p. 71. 

'" In wliat relates to the earliest times, antiipiities and history e.annot be en- 
tirely separated ; the comnienturii pontijicum and also Li^■y and Dionysius set us 
the example in this respect. — N. 


it is a fact which for the present requires no proof, that the 
Roman division of the nation into gentes answered to the 7€V77 
of the Greeks, and to the Geschlechter among our ancestors ; of 
this postulate the sequel of my exposition will furnish sufficient 
historical evidence. Let us first consider the nation respect- 
ing which we have more satisfactory information, I mean the 
Greeks. Their ^^evr] were associations which, notwithstandincr 
their common name, are not to be regarded as families, de- 
scended from the same ancestors, but as the descendants of those 
persons who, at the foundation of tlie state, became united 
into such a corporation. This is expressly stated by Pollux 
(undoubtedly on the authority of Aristotle), who says that the 
gennetae were named after the 76V7;, and that they were not 
united by common origin {yevei /xev ov TrpoaiJKovre'i), but by 
common religious observances {lepd). We, further, have the 
testimony of Harpocration respecting the Plomeridae in Chios ; 
for he says that they formed a 7eVo? which, according to the 
opmion of those learned in such matters, had no connection 
with Homer. These fyevr] moreover resemble the tribes of the 
Arabs: the Beni Tai are a body of 10,000 famihes, all of 
which cannot be descended from Edid Tai ; in like manner, 
the clans of the Highlanders of Scotland were named after 
individuals, but regarded themselves as their relatives and 
descendants only in a poetical sense: there were no fewer than 
5000 Campbells capable of bearing arms, who looked upon 
the Duke of Argyle as their cousin. 

With regard to the Eoman gentes we have no direct testi- 
mony like that of Pollux and Harpocration, that they were 
corporations without relationship; if we possessed Verrius 
Flaccus, we should undoubtedly learn something definite, but 
there is an important definition in Cicero^s Topica: he there 
mentions the term gentiles as a difficult term to define, and it 
had become so, because time had wrought various changes in 
the original constitution of the gentes ; in the tune of Cicero 
they had lost much of their former importance, and courts of 
justice had pronounced decisions respecting them. Cicero says : 
Gentiles sunt qui inter se eodem nomine sunt. Non satis est. Qui 
ab ingenuis oriundi sunt. Ne id qiiidem satis est. Quorum majormn 
nemo servitutem servivit. Abest etiam nunc. Qui capite non sunt 
deminuti. Hoc fortasse satis est. According to this, then, the 
Scipios and Sullas were gentiles^ for they were eodem nomine, etc. 


Supposing a Cornelius had been assigned as a nexus, or been 
condemned to death on account of some crime, he would there- 
by have ceased to be a member of his gens, and have incurred 
what the English in feudal language termed a corruption of 
Mood. If as an addictushe had children, they too were cut off, 
and did not belong to the gens. The addition quorum ma- 
jorum nemo servitutem servivii excludes all libertini and their 
descendants, although they bore the gentile name of their 
patronus ; but all peregrini might of course by common consent 
be admitted. The latter point, however, is probably an addi- 
tion which was foreign to the ancient gentile law ; for in my 
opinion there was at first no difference at all in regard to freed- 
men, they as well as the patrons belonged to the gens; but 
this was controverted, as we learn from the interesting suit of 
the patrician and plebeian Claudii (the IMarcelli) about the 
property of a deceased freedman." On that occasion^ it was 
a res judicata by the comitia of the centuries, that the patrician 
Claudii could not succeed to the property in dispute; whence 
was afterwards derived the doctrine that the libertini did not 
belong to the gens. 

Now in this definition there is not a single word about a 
common origin, a point which'could not have been over-looked ; 
and hence it follows that the Roman gentes were of the same 
nature as the Greek 'yevrj. Genus and gens are moreover quite 
the same word; similar variations often occur in the ancient 
language, as cliens and clientus^'^, Campans^^ and Campanus, 
and so also Romans and Romanus. The genitives Romanum 
and Romanom are formed from the old contracted nominative. 

It was a peculiarity of the institution of gentes, that the 
state was divided by legislation into a fixed number of associa- 
tions, each forming in itself a small state, with many peculiar 
rights ; it is possible that the expressions jus gentium and jura 
gentium originally signified something else, and something far 
more extensive than we understand by them. The number of 
the gentes is always found to bear so peculiar a relation to the 
state, that it can never have been the result of chance. In 
Attica there were 360, a number which the grammarians very 

" It is mentioned in Cicero, De Oratore. 

'* I have not been able to discover the form clientus, but the feminine clientu 
justifies us in assumini; the existence of a masculine in us. — En. 
" Nonius, 486. 24 ; Campus, Plant. Tri7i. ii. 4. 144 ed. Liud.— Ed. 


correctly refer to the division of the year or of the circle 
The same thing occiu's in Germany: at Cologne there were 
three orders, each containing fifteen gcntes; at Florence their 
number was thrice twenty-four, and in Dithmarsch thrice ten. 
Now at Rome there were probably thrice one hundred gentes, 
that is, three tribes each containing one hundred gentes, 
whence Livy calls them centuriae, not tribus. Between the 
division into tribes and that into gentes there usually existed 
another,which was called in Greece cjipaTpat, and at Rome curiae, 
answering to the orders at Cologne and to the classes in the 
Lombard towns. These curiae were parts of a tribe, but com- 
prised several gentes, probably always ten, for common religious 
purposes. As each gens had its own gentiUcian sacra — for 
sacra familiarum, which are sometimes mentioned by modern 
writers, did not exist among the Romans, — the membership 
of a curia implied special religious duties, and conferred the 
right of voting in the assemblies of the people. The ancients 
did not vote as individuals but as corporations, whence it was 
customary at Athens from the earliest times, to levy armies 
and to vote according to i^lnjlae (tribes) four of which might 
be out-voted by six, although the number of individuals con- 
tained in the six might be much smaller than that of the four. 
The Romans went even further, as they did not vote according 
to tribes but according to curiae, the reason evidently being that 
at first the Ramnes and Tities alone were the ruling citizens ; 
and to allow only these two to vote, would have given rise to 
difficulties, since it might easily have happened, that one tribe 
wished a thing which the other rejected, whereby collisions 
would have been produced. But as each tribe was subdivided 
into curies, and the votes were given according to this division, 
that difficulty was removed, and one curia might decide a 
question; this regulation therefore was necessary previously to 
the admission of the third tribe to a share in the government. 
At a later time, we find that the order in which the curiae 
voted and the praerogativa were determined by lot, an arrange- 
ment which cannot have existed at first, since the Lucercs as 
well as the two others might thereby have been chosen to strike 
the key-note. In this we have a glimpse of the innumerable 
stages through which the Roman constitution passed in its 
development ; and it was this very gradual development which 
secured so long a duration to Roman liberty. The secret of 


great statesmen, who are met with as rarely as any other kind 
of great men, is the gradual development and improvement of 
the several parts of an actual constitution ; they never attempt 
to raise an institution at once to perfection. 

Thus the curiae stepped into the place of the tribes. In the 
reign of Tarquinius, the third tribe, composed of the gentes 
mijiores, was admitted to the full franchise. The gentes are 
so essential a part of the constitution, that the expressions were 
gentes civium major es and minor es, just as gentes civium patriciac 
was the solemn expression for patricii. It is related that the 
senate, which till then had consisted of two hundred members, 
was increased by Tarquinius to three hundred by the admission 
of the gentes minores. This can mean nothing else than that 
he gave to the third tribe the full franchise, and admitted into 
the senate a number of persons corresponding to that of the 
gentes, for such is the natural course of things. At Cologne 
too, the second and third orders obtained access to offices later 
than the first. What Tarquinius did, was a great change in 
the constitution, which was thus completed for the first populiis. 
The third tribe, however, was not at once placed on a footing 
of perfect equality with the others, its senators being called 
upon to vote when those of the two other tribes had already 
done so; and there can be no doubt that their curies also were 
not permitted to vote imtil after the others. As regards the 
priestly offices, the members of the third tribe were admitted 
only to the college of the vestals. Wherever we find dmimviri, 
they must be regarded as the representatives of the first two 
tribes; triumviri do not occur till a later period, and wherever 
they are patricians, they represent the three tribes. They are, 
however, often plebeian, and in this case are connected with 
the plebeian constitution, which I shall describe afterwards. 


It is one of the most widely spread peculiarities of the earlier 
ages, and one of which traces have existed nearly down to our 
own days, that a distinction was made between the ancient and 
original citizens and those that were subsequently added 


to them. This distinction was inconsistent with the notions 
entertained in the eighteenth century, and has nearly every- 
where been abolished. In the United States of America the 
native population is extremely small; the office of presi- 
dent indeed can be filled only by a native, but in nearly every 
other respect it is perfectly indifferent how long a person 
has been in the country : and no distinction is made between 
the descendants of the first colonists and persons who have 
just settled there. In antiquity, on the other hand, admission 
to the franchise was every where more or less difficult, whe- 
ther the stranger spoke a different language or belonged 
to the same nation or even to the same tribe of the nation. 
In nations divided into castes, the admission is quite impossible, 
though the law is occasionally modified to favour a wealthy 
or powerful individual, as in the case of a Rajah who became a 
Brahmin on condition of his causing a colossal golden cow to be 
made, large enough to allow him to creep in at one end and out at 
the other. In some parts of the world, even at this day, a stranger 
is prevented from performing civil acts, and from obtaining offices. 
The earliest constitution concerning which we have authentic in- 
formation, though it is in part very obscure, is that of the Jews. 
They too had such a division ; the nation consisted of ten tribes 
with unequal rights, corresponding to the tribes of the Romans; 
beside them stood those who had been admitted into the com- 
munity of the Lord, that is the strangers. The Pentateuch 
expressly states that some nations were admissible, others not. 
The persons thus admitted into the community formed a 
multitude of people, who by religious consecration had become 
related to the Jews, but were neither contained in the tribes 
nor shared their rights. In later times, when the Jewish con- 
stitution becomes better known to us fforn contemporary records, 
the population is divided into Jews and Proselytes, and the 
latter again into Proselytes of righteousness and Proselytes of 
the gate.^ The former had politico-civil rights but were ex- 
cluded from civil honours; they might acquire land, make 
wills, many Jewesses and the like. The Proselytes of the 
gate were obliged to conform to the Jewish rites and were 

' These points connected with the second temple have been discussed by no 
one but tlic gi'cat Selden, without whom I sliould know nothing about them, since 
the Rabbinical language and literature are unknown to nie. bclden's reputation 
has very much decreased, at least in Germany; but it ought not to be so. — N. 


not allowed to act contrary to the ceremonial law, lest they 
might give offence to the Jews; but they did not participate 
in civil rights like the inhabitants of the country. 

The same institutions, though obscurely described, existed 
in all the Greek constitutions : much that is untenable has been 
written about them, but if once rightly understood they furnish 
a key to all ancient constitutions. In Greece, there existed 
from the earliest times, by the side of the sovereign body of 
citizens, an assembly of native freemen who enjoyed civil rights, 
but had not everywhere the connubium with the ruling people; 
they were protected by the state and might appear in the 
courts of justice, but had no share in the government. The 
condition of foreigners, freedmen and slaves, who had no 
civil rights was quite different, they being protected against 
injustice and oppression by taking a citizen as their guardian 
or patron. It was a very general notion that on the one hand 
a person might be a native and yet exercise civil rights only to 
a certain degree, and, on the other, that a stranger had no civil 
rights at all. 

The body of Koman citizens was now extended ; it was 
originally an aristocracy, only inasmuch as the subject people 
who lived in the neighbourhood stood to those citizens in the 
relation of clients, for otherwise no aristocratic relation is per- 
ceptible. But when Sabine and Latin communities became 
united with Rome in such a manner as to obtain full civil 
rights and to be obliged to serve in the armies, there arose a 
class of persons who, in our German cities, were called Pfahl- 
b'drger (Pale-burghers), an expression which no one has cor- 
rectly and clearly understood-. In Germany the word Paid 
or Pfnhl (Engl, pale ; in Ireland the counties about Dublin are 
said to be within the English pale) signified the district in the 
immediate vicinity of a city; the free people who inhabited it 
did not in reality possess the rights of burghers, which were 
peculiar to the gentes {Geschlechtei')^ but merely civil rights. 
The word was then gradually extended and applied to those 
strangers also, who attached themselves to a country or city 
(the Greek Isopolitcs). The investigation of this subject, 
Avhich is perfectly analogous to the origin of the Roman 
plebes, has given me much trouble, because in the sixteenth 

' Schiltcr on Konigshovcu has some good leniarks upon it.— N. 


century those relations died away, and no accounts^ of them 
are any where to be found. In the fifteenth century the word 
Pfaldbilrger still occurs; but in the sixteenth it is nearly obso- 
lete. J. V. JMuller did not understand it, and used it without 
attaching to it any definite idea. AVlaen a country district, or 
a town, or a knight, established such a connection with a city, 
two consequences followed; first they mutually protected one 
anotlier in their feuds, and the strangers with their vassals 
might remove to the city where their civil rights were perfect- 
ly free, and where they also had their own courts of justice; 
but they did not form part of the ruling body ; and in this 
respect they were distinguished from the gentes or Geschlechter, 
who exercised the sovereignty. Many Transtiberine commu- 
nities, both Latin and Sabine, entered into this relation with 
Rome, and formed settlements, especially on the Aventine. In 
describing this, the Roman historians speak as if Ancus had 
removed those people from their homes and given them settle- 
ments at Rome, a state of things which is inconceivable ; ;. for 
all the country around Rome was previously occupied, so that 
there they could not settle, and therefore they would have 
been obliged to take up their abodes at a distance of many 
miles from their fields. It is very possible, however, that a 
few of the highest rank were obliged to settle at Rome, 

This pale-burghers' right was extended further and further: 
the multitude which enjoyed it did not yet form a corporation, 
but contained all the elements of one ; they became so nume- 
rous at Rome and in the surrounding country, especially 
through the alliance with Latium under Servius Tullius, that 
the pale-burghers far surpassed the ancient population in 
numbers, formed the main strength of Rome, and were especi- 
ally employed in war. With their increase, the decrease of 
the burghers who married only among themselves kept pace. 

In this manner arose the Roman plebes, in Greek Sfj/xo^, 
and, as we call it, the commonalty. The demos comprised all 
those who had the lower franchise, and therefore owed oblio-a- 
tions to the state, but had no rights except their personal 
freedom. Thus the same relation is expressed by the words 
S^/A09 and TToXirai, as by plehes and populus, or commonalfij and 
burghers^ or lastly commune and cittadini? I further believe 

' These relations were so familiar to our ancestors, that in the old translation 
of Livy published atMaycnce,/JO/)«/«s is throughout translated by Gcschlechter , 


that originally tlic city was not called TroXt? but acrrv : TrdXt? like 
populus is a Tyrrhenian word, and both have the same mean- 
ing, populus being formed by reduplication from ttoXi?- The 
commonalty was the principal part of the population in all 
states as far as numbers are concerned; but its development 
did not take place in antiquity in the same manner as in the 
middle ages. In the latter, the commonalty lived within the 
walls of a city ; and they often, as was the case at Geneva, 
settled around the city {cite or the nucleus of a town) , in what 
was called bourg, horgo or suburbs, and were thence called 
bourgeois. These suburbs in the course of time were fortified 
and obtained equal rights with the cities. In Germany the 
case was the same, the name only being different, for burghers 
and Geschlechterare identical, and towns were formed, especial- 
ly after the tenth century, when peace had been restored to 
the world. "Wherever in Gaul a civitas existed from the time 
of the Eomans, it was called a cite; and where there was a 
royal villa, it often happened that a place sprung up in the 
vicinity under the protection of the king, and under the ad- 
ministration of the king's major domus. This is the original 
meaning of ville, as contradistinguished from cite. Hence in 
French towns a distinction is made between la cite, la ville 
and le bourg. Where the commonalty sprang up within the 
walls, it had quite difterent elements. Throughout the Ger- 
manic states, strangers were, on the whole, more kindly 
treated than in ancient times or in France. The free settlers 
in the small Swiss cantons, as inUri for example, were in reality 
oppressed commonalties; the inhabitants of St. Gervais were 
subjects of Geneva. Among the Slavonic nations, as at Novo- 
gorod, such settlers were called guests, and their condition was 
in many respects easier than that of the natives. In France, 
down to the time of the revolution, strangers were not able to 
make a will, and according to the droit d'Aubaine, the sove- 
reign succeeded to their property if they were not naturalised. 
The same law also existed in England, where to this day 

and plehes l>y commonaUy. There we meet with expressions such as this: 
" T. Quinctius was elected burgomaster from the Geschkchfer and L. Genuciiis 
from the commonalti/," where Livy lias populus and pJebes. Tiiis iinsopliisticatcd 
way of viewing things is the reason why the men of the sixteenth century, 
though without the learning wliich we require, yet comprehended many things 
quite correctly. It is only a few weeks since I found tin's out. — N. 


foreigners cannot acquire landed property. In all the towns 
of the middle ages in which commerce was the principal 
occupation, the commonalty soon formed itself into guilds, 
which obtained their own presidents, and masters of the guild, 
as well as their own laws and courts: penal jurisdiction could 
be granted by the kings alone, and wherever it was exercised 
the guilds took part in it. The masters of the guilds at first 
appeared in the council only for the purpose of taking care 
that their rights were not trespassed upon; but they soon 
became members of the council and finally obtained the upper 
hand. This is clearly seen in the Italian towns, as, for in- 
stance, in the seven ancient guilds at Florence. Durino- the 
feuds of the Guelfs and Ghibellines, the burghers were still 
the masters; but soon after, about the time of Eudolph of 
Hapsburg, the guilds everywhere had the ascendancy, in Italy 
in the thirteenth, and in Germany about the middle of the 
fourteenth century^ as at Zurich, Augsburg, Strasburg, Ulm, 
Heilbronn, and the imperial cities of Suabia. During the 
period of transition, the burghers shared the government with 
the guilds ; wherever this was done , the union was brought 
about peaceably; but where the burghers refused, it was 
effected by a bloody contest, which mostly ended in the 
destruction of the burghers, though the case was sometimes 
reversed, as at Nlirnberg, where the guilds were oppressed. 

The union of the burghers and the commonalty or guilds 
was called in Greece iroXcreia, in Italy popolo, the meaning of 
which is somewhat different from the Roman populvs^ The 
distinction between the burghers and the commonalty went 
so far, that at Florence, for example^ in the palazzo vecchio, 
and also on books, one sees a lily as the armorial bearing of the 
city, by the side of a red cross on a silver ground as that of 
the commonalty iil commune). The expression il commune 
may very easily mislead; it does not denote the union between 
the two orders, but the commonalty, a fact to Avhich Savigny 
has directed my attention; at Bologna there is a palatkan 
civium and a palatium communis. The Capitano del popolo and 
the Capitano di parte at Florence are likewise difficult to un- 
derstand. During the struggle between the Guelfs and 
Ghibellines, the Capitano di parte, that is, of the Guelfian 

■• The investigations into the histoiy of the Italian towns which I liave made, 
throw great h'ght upon the whole development of the Roman constitution. — N. 


party, drove the Gliibcllines from the city : he was placed at 
the head of affairs, and the franchise of the others was suspend- 
ed. The only Capitano of the burghers was now nevertheless 
called di parte. 

Among the ancients, on the other hand, it was not the 
guilds within the walls that formed the commonalty, but the 
inhabitants of the country around the city, which consisted of 
different elements and embraced both the noblest and the low- 
est. It is therefore a most preposterous notion, that the 
plebes consisted of the poorer classes only. This error was 
caused by the imperfection of the language, such as it appears 
even in the writings of Plato and Aristotle, for the Greeks 
had only one word hr)p,o<i to designate the burghers, the com- 
monalty, the union of both and, in short, the whole people as 
well as the populace, in contradistinction to the rulers. 
Dionysius knew the word Sr)fio<i only as opposed to /SouA,^, 
and 6)(Xo<i is the proper term for the mass of poor people. 
But even he is not free from misconception, which he trans- 
ferred to Roman history, and as he is much more minute than 
Livy, in describing these relations, he has led the restorers of 
ancient history to adopt quite erroneous notions. Livy too 
does not see clearly into the matter, but he has many passages, 
from which it is evident that the annalists whom he followed 
had taken the right view. A further cause of confusion 
arose from the distress and debts which are often mentioned 
as prevailing among the plebes, which, however, as we shall 
hereafter sec, must be referred to debts arising solely from 
mortgages of landed proprietors. The plebes was distinct 
from and opposed to the populus ; the Romans in general 
divided all the fundamental powers in nature, as well as in the 
realm of spirits, into two parts, describing them as male and 
female; for example Vulcan and Vesta are fire, Janus and 
Jana the heavenly lights of sun and moon, Saturn and Ops 
the creative power of the earth, Tcllumo and Tellus the earth 
as firm ground ; and in like manner, the complete state con- 
sists o^ populus and plebes, which together constitute the whole. 

Within the territory of the ancient city, which extended 
about five miles on the road towards Alba, and the limits of 
which can be very accurately fixed, there ^ lived under the 

* I am sorry thdt I did not find this out while I waf5 in Italy, for I had often 
been where that limit must have existed, without noticing it. It was not till 


protection of the populus a number of clients {cluentes, from 
cluere, to listen). It was owing to a great variety of circum- 
stances, tliat these clients came to be connected with their 
patrons, in the same manner as vassals were with their feudal 
lords, so as to be obliged to ransom them from captivity, to 
provide dowries for their daughters, and to defend them in all 
cases of need and danger. Some of them may have been 
ancient native Siculians, who being subdued by the Cascans 
undertook those feudal obligations in order that their lives 
might be spared; strangers may have settled in the Eoman 
territory as aliens and have chosen a Eoman citizen as their 
guardian ; some also may have been inhabitants of those places 
which were obliged to take refuge under the supremacy of 
Rome; slaves lastly who received their freedom stood to their 
former masters in the relation of clients. This class of per- 
sons must have been ever on the increase, so long as Rome was 
in a flourishing condition. The asylum in the ancient tradition 
must be referred to the clientela, for the clients had actually 
come together from all parts. But the free commonalties 
inhabiting the country districts were quite different: their 
origin was traced to the times of Aucus. Scaliger, by one of 
the most brilliant discoveries, found out that Catullus calls the 
Romans ffens Romulique Ancique, where Romulus represents 
the burghers and Ancus the commonalty. The plebes now 
gradually increased, partly by the extension of the Roman 
dominion, and partly by the circumstance that, when a family 
of buro-hers became extinct and its former clients were "svithout 
a feudal lord, they attached themselves to the commonalty; 
many also joined the plebes in consequence of the alliances of 
Rome with free towns. Such relations, however, are in their 
origin imperfect, but become more and more clearly developed 
in the course of time: at first they were entirely local, and 
es like Tellcne, Ficana, and Politorum, were undoubtedly 
at first quite isolated and without any regularly organised power. 
There can be no doubt, that a populus and a plebes existed in all 
the towns of Italy and also in the Greek colonies of southern 
Italy and Sicily, the constitution of which bears a strong- 
resemblance to that of the Italian states, and sometimes even 
adopted the same names. 

last year that by a simple combination and with the assistance of Fabretti's map 
of the neighbom-hood of Home I made the discoveiy. — N. 
VOL. I. G 



Previously to the time of Servius Tullius, tlie country about 
Rome was not united with the state, at least probably united 
only tlirough the king, that is, the inhabitants were obliged to 
obey the government, but were otherwise treated as perfect 
strangers ; they did not even possess the commercium, that is, 
no patrician could acquire landed property in the country 
districts any more than a plebeian could at Rome. The same 
regulation has existed in many countries down to recent times, 
so that the landed property of a peasant could never be acquired 
by a nobleman: a very wise and salutary regulation, which 
unfortunately has been abolished, in consequence of the 
erroneous belief that It was a foolish restriction. It is still 
less conceivable that the plebeians should have possessed the 
legal right of contracting marriages with the patricians; the 
children of such marriages In all cases followed the baser side. 
The Mensian law^ did not Invent this, but was merely a re- 
enactment, determining more minutely what was to be done 
in difficult cases. But there now appeared a legislator, who, 
on the one hand, gave to the commonalty a constitution which 
was complete in Itself, and, on the other, devised forms by 
which this commonalty became united with the whole body of 
burghers. The former part of his legislation has been entirely 
overlooked, and the latter appeared quite mysterious to Livy 
and DIonysIus; so great had been the change of aifalrs since 
the days of Fabius, who still had a correct view of these 
matters, though only two hundred years had elapsed from his 
time. Let him who thinks that this Is impossible, look around 
himself: I believe that In this town [Bonn] there are not three, 
and at Cologne not ten persons, who can state precisely what 
the, constitutions of these towns were two or three hundred 
years ago, nay, not even what they were previously to the 
year 1794. Of this fact I satisfied myself In 1808, In con- 
versation with a Frieslander who had devoted himself to 
historical pursuits, but was unable to give me any account of 
the constitution of his country before the French revolution. 
The same is the case at Brussels. In countries where the 

' Ulpian, Fragm. v. 8. -Ed. 


constitution has been as little changed as in England, it is 
easier to trace one's way back from the present to the past. It 
is scarcely credible how great a change two hundred years 
may bring about, and how distant the whole mode of thinking 
and living seems to be, when separated from us by some great 
event. Such was the case in Germany after the seven years' war : 
all German literature previous to that event presents to our 
minds a character of strangeness, whereas that of the period 
immediately succeeding seems to us as if it were more or less 
of yesterday. Such a crisis in literature and in the entire mode 
of thinking had taken place at Rome through the influence of 
Cicero; so that Livy, Virgil, and Horace, must have thought 
the authors of the preceding period as strange as we think 
those who wrote before Lessing and Goethe. The Julian law 
likewise had so completely changed many circumstances in the 
civil rights of the Latin allies, that the recollection of the 
preceding state of things was entirely obliterated. The new 
constitution was simple, and the ancient complicated institutions 
were no longer intelligible. Thus it becomes evident — and 
I beg of you to mark this well — that even ingenious and 
learned men like Livy and Dionysius did not comprehend the 
ancient institutions, and yet have preserved a number of ex- 
pressions from their predecessors, from which we, with much 
labour and difficulty, may elicit the truth. 

The statement of Dionysius, derived from Fabius, that 
Servius divided the city and the country, forming the territory 
of Rome into thirty tribes, is an instance of what I mean. The 
division of such a territory was topical : it was not a peculia- 
rity of the Romans, V but is also found in Greece, where 
Cleisthenes took the ager Atticus as the basis for the division of 
the Attic nation. The whole was divided into a fixed number 
of parts; and in order to eifcct this, the legislator did not 
count the large towns, but took a convenient number, such as 
one hundred, into which the country was to be divided, so 
that some large places were cut up into parts, while smaller 
ones were combined into one. These divisions according to a 
fixed number were so universal among the Romans, that when 
Augustus divided the city into fourteen regions he did not 
count the vici^ but assigned a definite number of vici to each 
region. Now the legislator whom we call Servius TuUius 
divided the city of Rome in so far as it was inhabited by pale- 

G 2 


burgliers, into four, and tlie territory around it into twenty-six 
regions. This must be looked upon as true : but to prove that 
this statement of Fabius is correct would lead me too far. 
Here it must be observed, that the existence of a populns nearly 
always presupposes the existence of a plebes as its counterpart, 
and accordingly a plebes, though unimportant, must have ex- 
isted even before the time of Ancus. Each of the three towns, 
Roma, Quirium, and Lucerum, had its own commonalty; these 
commonalties and the settlers on the Esquiliae under Ancus 
form the four city tribes; the first or Palatina corresponds to 
the Palatine, the second or Collina to the Quirinal, the third or 
Suburana to the Caelius, tlie Carinae and Subura, and the 
fourth or Esquilina to the Esquiline and Viminal. This 
arrangement must have been made before the building of the 
wall of Servius Tullius, as is clear from the existence of the 
Esquilina. The division was purely geographical, and not at 
all connected with certain families ; the territory was the basis, 
60 that the inhabitants of a certain district formed an associa- 
tion of peasants (Bauernschaften). It cannot surprise us to 
find such associations of peasants within the city, for at Ant- 
werp some of the streets of the extended city are still called by 
a name {Burschafteii) which indicates that originally they were 
inhabited by associations of peasants which formed themselves 
by the side of the ancient city. Such a division resembles 
our political divisions based on locality and domicile, but there 
is this diiference, that ours are not permanent: so long as, e.g. 
I live at Bonn, I am a citizen of Bonn, but I should cease to 
be so if I were to remove to Cologne. When this division 
was made at Rome, every one received a name from the region 
in which he lived, but when he changed his abode he did not 
thereby cease to belong to the local tribe corresponding to the 
region in which he and his descendants were registered. I do 
not mean to say that a change was impossible, but all important 
changes belong to a time when the tribes had acquired quite 
a different and much greater importance than they had at first.' 
During the first generation, matters may have remained as they 
were established by the legislator, but in the course of time 
changes m.ust have taken place, as people did not always 
continue to reside in the same district. 

' In the canton Schwyz, likewise, the country people were di^ndcd into four 
quarters, in which they were enrolled and of which they remained members 
although they might take up their abode in another quarter. — N. 


The names of the country tribes were originally derived not 
from the districts but from heroes, who were eponymes both for 
the tribes and the burghers; for it is evidently the object of 
this legislation to amalgamate the different elements of the 
people ; and the recollection of former times, when those places 
had been independent, was to be effaced by the thought that 
they were Eomans. They obtained common sacra like the 
tribes of the burghers, as is expressly mentioned by Dionj^sius, 
for in antiquity sacred rites were always a bond of union. The 
fact of the plebeian tribes having sacra is also established by 
the circumstance, that Tarqiiinius Superbus expressly forbade 
them. Every tribe or region in the city was subdivided into 
vici and those of the country into pagi, and each of these vici 
or pagi had its own magistrate^ as every tribe had its trihunus. 
Reo-ulations of the same kind were in force at Athens; when, 
for example, a person was enrolled at Acharne and removed 
to Sunium, he still remained an Acharnian. As these tribes 
in the earliest times all possessed equal privileges, there was no 
motive for wishing to be enrolled in another tribe; but after- 
wards when there arose a difference of political rank among 
the tribes, of which I shall speak hereafter, matters were 
changed; the city tribes became inferior to the country tribes, 
and to be removed from the latter to the former was a nota 
ignonmiiae, a practice which may be dated from the censorship 
of Fabius Maximus. The tribes contained only plebeians, the 
patricians being comprised in the curies which also included 
their clients, ^^^len a person became a Roman citizen with- 
out the suffrage, he was not received into a plebeian tribe, nor 
was it possible to be admitted by isopolity or by manvimission, 
and consequently he could not be invested with any office, nor 
vote in the assembly. The qualification for voting in a ple- 
beian tribe consisted in being a landed proprietor and agri- 
culturist; whoever supported himself by any other occupation 
was excluded. 

In this manner the legislator constituted the two corporations 
of the patricians and plebeians : he might have united them in 
two assemblies, as in modern states, but this was impracticable 
in those early times, as the two corporations regarded each 
other with hostile feelings. In order to effect an accommoda- 
tion, Servius created the centuries, like the concilio grande at 
Venice, in which, as soon as they entered the hall, all were 


equal, poor or rich, every one being in simple attire. The 
object of the centuries was to unite the patricians and plebeians, 
as well as those who sprang up by the side of the latter and 
occupied their former position ; and at the same time to exclude 
those who had no landed property, and could therefore give no 
guarantees to the state. The centuries accordingly contained 
the whole of the first estate ; of the second, those who had the 
right of voting; of the third, those whose property was equal 
to that of the second ; and lastly, persons engaged in certain 
honourable occupations. The statements of Livy and Diony- 
sius have caused great confusion in this part of Koman history, 
as they conceived the tribes differed only in rank and property ; 
they believed that the old citizens, that is the patricians, were 
divided into curies and were perfectly equal among themselves, 
but they imagine that this was an oppressive democracy which 
Servius Tullius abolished by the introduction of the centuries. 
It is the same error as that into which Sismondi has fallen, 
who fancies that the Italian towns, on their first appearance in 
history, were under a democratic government: a monstrous 
mistake ! Had the Roman historians attentively studied the 
ancient law-books, these things certainly could not have re- 
mained obscure to them ; but after all, we ourselves have not 
fared bettcri, for it is now scarcely fifty years since Moser pub- 
lished his first works, stimulated by which we have at length 
begun to have a clear perception of the early institutions of 
our own country. 


According to the primitive institutions, the burghers' served 
not only on horseback, as was the case afterwards, but also on 
foot; the same was originally the case in the German cities. 
These burghers at first had nothing in common with a nobility. 
We may assume that each gens furnished one horseman and 
ten foot soldiers; hence the statement in Plutarch that the city 

' The German word here is ein Gescldcrhtci, which in early times, as in the 
Chroniule of Cologne, denotes a person belonging to a Geschkchl. — N. 


at first consisted of about a thousand families. This looks very 
historical, but such additions, as about and the like, in Plutarch, 
Dionysius and other writers of later times, are meant as soft- 
eners of colours which appear to them too glaring ; the 
statement is indeed very ancient, but is a symbolical representa- 
tion of a legal relation rather than an historical fact. Rome 
in the earliest times contained one hundred gentes, and conse- 
quently one thousand foot soldiers, each of whom was consi- 
dered to have been furnished by a family.^ Along with 
these the country districts sent their contingents, which were 
probably levied according to the townships. The new legisla- 
tion reformed the phalanx, exempted the burghers from the 
obligation to serve on foot, and made them serve on horseback 
with particular privileges. As the whole burthen of forming 
the infantry now devolved upon the commonalty or plebeians, 
corresponding privileges were granted to them, and thereby 
also the means of maintaining their freedom. Thus the popu- 
lation was divided into cavalry and infantry, the commonalty, 
however, not being excluded from the former. The infantry 
of all European nations in ancient times resembled the Greek 
phalanx. It was a mass which produced its effects by its irre- 
sistible onset: the men were armed with pikes, with which 
they advanced against the enemy in eight, ten, or twelve ranks. 
Barbarians did not fight in close masses, and the Asiatics were 
only archers. When, as at Rome, the soldiers were drawn up 
ten men deep, those in the rear were of course less exposed 
and did not require the same protection as those in front: 
when they properly closed their shields they needed no coat of 
mail, and the last rank not even greaves. Some also were 
light troops or slingers, who threw lead and stones. Every 
one in the infantry was obliged to equip himself at his own 
expense and in proportion to his property, the wealthier hav- 
ing to provide themselves with full armour, while those of 
small means were only required to serve as slingers. When 
a war was protracted, gaps arose, and after an unsuccessful bat- 
tle, the first lines might be much thinned, so that a complement 
became necessary: in such circumstances those standing be- 
hind put on the armour of the slain, and stepped into their 
places. In all campaigns, however, there was also a reserve in 
case of need. These were the three elements of the Roman 
^ I have neglected to explain this in my history. — N. 


army: the legion properly so called, the light-armed, and lastly 
the reserve, which took the place of those who had advanced 
from the hindmost lines to supply the place of those who had 
fallen in front. 

Servius thus regarded the whole nation, populus and plebes, 
as an army, exercitus vocatus ; but when this army marched 
against an enemy, it further required carpenters to build 
bridges, erect tents, and the hke, and musicians; the former 
were constituted as one, and the latter as two centuries; and 
this addition really completed the army or classis.^ These three 
centuries did not consist of plebeians, for no plebeian was al- 
lowed to engage in any other occupation than agriculture; if 
he did, he renounced his order, and the censors erased his 
name from his tribe (capitis deminutiu), which, however, was 
not originally attended with any disgrace. There existed at 
Rome from the earliest time certain guilds, the institution of 
which was ascribed to Numa: their number was three times 
three, pipers, goldsmiths, carpenters, dyers, saddlers, tanners, 
coppersmiths, potters; and the ninth included all other kinds of 
artificers. The object of this undoubtedly was, to give to the 
city trades a corporative existence, as in the middle ages; but, 
as the persons contained in these centuries were usually freed- 
men and foreigners, the object of whose ambition was to quit 
these associations and become enrolled in a tribe, the guilds 
never attained any high degree of prosperity. At Corinth 
they were of greater importance. By this division into centu- 
ries, the plebeians were connected both with the patricians and 
the aerarians ; carpenters and musicians, who were of so much 
consequence in war, had special centuries assigned to them, 
whereby they obtained the same rights as would have belonged 
to them if they had served in the army as plebeians. The 
carpenters, in consideration of their importance, were ranked 
with the first class, and the musicians with the fifth. 

Lastly Servius also took notice of those free people who did 
not belong to the commonalty. Many of them undoubtedly 
entered the service either by compulsion or of their own accord ; 

' In the account of the buttle of Fidcnac, Livy is much puzzled by this word : 
the ancient annalist had the classibus certare, which Li\'y mistook for 
fleets, and hence he expresses a doubt as to the possibility of an engagement be- 
tween two fleets in the narrow river Tiber; but the phrase merely meantabattlo 
between two armies in full armour. — N. 


for I cannot believe that the capite censi and the proletarii did 
not perform any service at all ; they did not fight against the 
enemy, but served only in the baggage train, as lixae and 
calories, who there is no reason for supposing were always 

Servius thus had a perfectly organised army, which with the 
addition of the cavalry he made the representative of the nation. 
He composed the cavalry of the three ancient double tribes or 
six centuries of Tarquinius Priscus, and to them he added 
twelve other centuries of the plebes, consisting of the most 
distinguished persons of the commonalty. Those six centuries 
comprised the entire patrician order, which on the whole cer- 
tainly had a small number of votes, but as we shall hereafter 
see, it had a preponderance in other respects: among them 
there was perfect equahty, and no difference was made on 
account of age, each century having one vote. Within the 
plebeian order Servius Tullius separated the more noble and 
wealthy into two classes, the first consisting of those who had 
formerly belonged to the Latin nobility, and the second of 
those who had not. To the class of nobles he assigned the 
twelve remaining equestrian centuries, and this without any 
regard to their property, except that those who had become 
quite impoverished were probably omitted. This is a point 
which you must bear in mind ; for, according to the prevalent 
opinion based upon an incorrect expression of Cicero {censu 
maximo), the members of these twelve centuries are said to have 
been the wealthiest among the plebeians. Had the equiies been 
the wealthiest then as they were after the Hannibalian war, how 
senseless would the constitution have been ! There would have 
been no division of property between 1,000,000 sesterces, the 
sum fixed for this class after the Hannibalian war, and 100.000; 
whereas, from the latter sum downwards, there appear a num- 
ber of divisions. AVe have moreover the express testimony of 
Polybius, that the property quahfication of the equites was 
something new and opposed to the ancient notions, according 
to which, descent was the determining point. Lastly, another 
proof is contained in the testimony that the censors could dis- 
tinguish a plebeian by enrolling him among the equites, a fact 
which excludes classification according to property. Under 
Augustus, things certainly were difterent; for at that time the 
most distinguished men could not become equites without a 
certain amount of property. 


Now what is to be understood hy census F Among ourselves, 
every kind of property and all rights which can be estimated 
in money would be included in it. But among the Eomans it 
was different; and it must be regarded as an undoubted, fact, 
that the census affected only res corporales, that is substantial 
objects, and not res incorporales, such as debts. If, for example, 
1 have a piece of land worth fifty thousand asses, and owe ten 
thousand to another person, my property in reality amounts to 
only forty thousand asses ; but such things were not taken into 
account in the census of the ancients, and debts were not 
noticed at all. This very important and decisive point has not 
been attended to by the earlier writers on Roman history, be- 
cause they were not men of business. We must not regard the 
census as a property-tax, but as a land-tax or a complex of 
direct taxes: certain objects were estimated according to pre- 
scribed formulae, at a particular value; and a certain per- 
centage was paid on that estimate. In the Dutch part of 
Friesland, lands were valued in pounds, and upon these pounds 
a certain tax was levied; hence a piece of land was called pon- 
demate. The Roman census then comprised all property in 
land, and undoubtedly also all res mancipi ; but I am convinced 
that nothing was paid on outstanding debts, even though they 
might constitute the property of the richest man at Rome. 
The Attic census, on the other hand, was a real property-tax. 
The consequence was, that at Rome the whole mass of move- 
able property possessed very little influence ; for the wealthiest 
capitalist might be entirely free from taxes, landed property 
having to bear all the burthens, but at the same time enjoying 
all the privileges: in this point the census accurately corre- 
sponds to our direct taxes, in imposing which likewise no notice 
is taken of any debts with which the property in land may be 

All those Romans who were not contained in the equestrian 
centuries, were divided into such as possessed more than 12,500 
asses, and those whose census did not come up to that sura. 
The former were subdivided into five classes; among them 
were no patricians, but all those plebeians whose census amount- 
ed to the specified sum, and the aerarians, that is, those who 
were not contained in the tribes, but whose property placed 
them on an equality with them ; the aerarians were now what 
the plebeians had been before, and, if they acquired landed 


property, they were enrolled in a tribe. The first class com- 
prised all those who possessed 100,000 asses or upwards, and 
their property might consist of land, metal, agricultural imple- 
ments, slaves, cattle, horses and the like : it was divided into 
eighty centuries. All persons from the age of sixteen to forty- 
five were counted sajuniores, those from forty-five to sixty as 
seniores. At Sparta a man was liable to serve in the army till 
his sixtieth year; but at Rome, the seniores had no other duty 
than to defend the walls of the city. The seniores undoubtedly 
did not form one half of the whole population ; for under the 
favourable circumstances of a southern climate, thay could 
hardly have amounted to more than one third or more accu- 
rately to two sevenths ; all persons alive above the age of forty- 
six may perhaps have been no more than one half the number 
of the juniores. There is every probability that at that time 
all civil rights and civil duties ceased with the sixtieth year. 
In Greece^ a higher value was set upon the abilities of old age ; 
among the Melians, the whole government was entrusted to 
the hands of the old men above sixty. Although the seniores 
at Eome were in number only about half as many as the 
juniores, yet they had an equal number of votes with them, 
and probably voted first. 

The remaining four classes were valued at 75,000, 50,000, 
25,000 and 12,500 asses respectively. The second, third, and 
fourth, had each twenty, and the fifth, thirty centuries. One 
hundred thousand asses were not a large property, being about 
the same value as 10,000 drachmae at Athens, one as being 
about 3^ farthings English. In the army, each century served 
in a fixed proportion, so that a century which contained a 
smaller number of citizens performed a greater proportionate 
amount of military service than the more numerous ones. It 
was a combined levy from the tribes and the centuries. Within 
the thirty tribes, one man was always called up from each 
century of the juniores, so that each century fiirnishcd thirty 
men. Each succeeding class had to furnish a greater number 
of troops, in such a manner, that while the first furnished a 
single contingent, the second and third had each to furnish a 
double one, the fourth a single one, employed as dartsmen, 
and the fifth again served with a double contingent. 

The object of the constitution, based as it was upon property, 
would have been completely lost, unless the first class had 


had a preponderance of votes. The centuries in the lower 
classes became larger in the number of persons contained in 
them, in proportion as their property decreased, so that of 
thirty-five citizens possessing the right of voting six only 
belonged to the first class. Dionysius is here perplexed in the 
detail, but he ^had before him a distinct statement that the 
summing iip was made according to property. 

All those whose taxable property did not amount to 12,500 
asses were again sub-divided into two sections: those who 
possessed more than 1500 asses still belonged to the locnpletes ; 
those who had less were called joro/e/ffriz, that is persons exempt 
from taxes : they formed one century. The locupletes embraced 
all the plebeians except the proletarii, and were so far quite 
equal among themselves; but between them and the proletarii 
there was a gulf; any locuples, for example, might in a court of 
justice be surety for another person, but not so a, jiroletarian : 
it is clear that those only could be vindices with sums of money, 
Avho could prove from the registers of the censors that they 
possessed such money; there is moreover no doubt that only 
locupletes could be chosen by the praetor as judices, or come 
forward as witnesses, as is proved by the expression locupletes 
testes. The proletarians therefore belonged to quite a different 
category, but whether they were at that time allowed to vote in 
the plebeian tribes is uncertain. 

Such was the constitution of the centuries of Servius, re- 
specting which Livy differs from Dionysius, and both again 
from Cicero's statement in the second book ])e Re Publica; 
but this passage though very corrupt may be emended. The 
sum total is 195 centuries, of which 170 belonged to the five 
classes, two of the locupletes or assidui (the accensi and velati), 
two of the proletarians (the proletarii in a narrower sense, and 
the capite censi), the three centuries of the trades ; and lastly 
the eighteen equestrian centuries, six consisting of patricians 
and twelve of plebeians. The passage of Cicero has given 
rise to several conjectures, all of which are erroneous, as, for 
example, that of the celebrated Hermann; but if a person is 
familiar with such investigations, all may be made clear by 
the Roman numerical combinations, which I have developed, 
the object of the whole institution was, that the minority should 
have a decisive influence'*, wealth and birth having all the 
* The Abbe Sieyes, it is true, has said, la minority a toujours tort.—H. 


power; for the eighteen equestrian centuries and the eighty 
centuries of the first class were first called upon to vote ; if they 
agreed on any question it was decided at once, as they formed the 
majority of centuries, though they contained by far the smaller 
number of citizens. Among persons of the same class again it 
was the minority which decided, because the forty centuries of 
the seniores contained far fewer voters than the juniores. 

If this institution had had the meaning assigned to it by the 
historians, it would have been highly unjust towards the patri- 
cians, who surely still formed a considerable part of the nation. 
These historians did not see that the patricians did not belong 
to the classes at all — their presence in the centuries being only 
a representation, and consequently only of symbolical import- 
ance — but they merely said that the patricians probably voted 
with the wealthy, that is in the first class; now the patricians 
were by no means wealthy according to the census, since they 
possessed the floating capital only, not the allodia. But the 
alleged injustice did not exist, for the centuries stood to the 
curies in the same relation as the House of Commons stands to 
the House of Lords. No election nor law was valid, unless 
when sanctioned by the curies, which sanction is implied in 
the expression ut p aires auctores fierent ; the centuries moreover 
could not deliberate on any subject which had not been pro- 
posed by the senate, and no member of a century had the right 
to come forward and speak ; which right was certainly possessed 
by the members of the curies. In the assemblies of the 
tribes, the discussion of subjects proposed by the tribunes 
seems indeed to have been permitted, until the voteswere taken ; 
but this permission was probably not often made use of The 
power of the commonalty in the centuries was thus extremely 
limited; it was merely one step towards republican freedom. 
At that time the assembly of the tribes had nothing to do with 
the framing of laws; they could only elect their own officers 
and make arrangements concerning their local interests; three 
may have been among them regulations respecting the poor, 
for bread was distributed under the superintendence of their 
aediles at the temple of Ceres; but their most important power 
was conferred upon them by Servius Tullius,who granted to the 
plebeians the right of appeal to the assembly of the tribes 
against sentences of punishment pronounced by a magistrate 
upon disobedient individuals. The privilege of an appeal to 
tlie cii ries had long been possessed by the patricians. 


The laws of Servius Tullius may have contained far more than 
we know, but Tarquinius Superbus is said to have completely 
abolished them, that is, they were not found in the jws Papiri- 
anum. It is stated that there were fifty laws. How far the 
equalisation of the two estates was carried is uncertain ; but the 
exclusive right of the patricians to the domain land, and the 
pledging of a creditor's person are said to have been abolished. 
It is more certain that the legislator intended to lay down the 
kingly dignity and to introduce the consulship in its stead, so 
that the populus and plebes should each be represented by a 
consul, an idea which was not realised till one hundred and 
fifty years later by the Licinian law. Servius looked upon 
himself as a vofMoOirr]'; like Lycurgus or Solon. This change in 
the form of government would have been easy, for the kings 
themselves were only "magistrates elected for life, like the stadt- 
holder in Holland, or the President in the United States, who 
is elected for four years ; and such constitutions seem to have 
been very frequent among the early Italian nations. The elec-' 
tion of two consuls appears to have been prescribed in the com- 
mentaries of Servius Tullius^ ; but it was not carried into effect, 
either because his life was taken away too early or because he 
himself deferred it. Tanaquil is said to have entreated him not 
to renounce the throne nor to forsake her and hers. What is 
ascribed to Servius Tullius was not entirely accomplished by 
this king, but occasioned the revolution of Tarquinius Super- 
bus. Although Servius is stated to have reigned forty-four 
years, still Livy mentions only one war, that against Caere 
and Tarquinii, which was brought to a close in four weeks. 
Dionysius, too, relates no particulars that have even an appear- 
ance of truth. The time of his reign is much too long in our 
accounts, and it was probably very short. 

The same legislator is said to have permanently settled the 
relations between Rome and the Latins. The report is, that 
he concluded an alliance with the latter and induced them to 
erect a common sanctuary on the Aventine, in which the 
tables of the league were set up, and in which Rome offered a 
sacrifice, a circumstance which, as Livy says, was a confessio 
rem Romanam esse superiorem. The investigation into the con- 
dition of the Latin people is one of the most difficult : at first 
every thing seemed to me to be a mass of confusion, and it 

* lAxy says: duo consules creati sunt ex commentariis Servii Tullii. — N. 


was only step by step that I began to see clearly. It is a 
mistake of the ancients, which I shared with them till very 
recently, that Servius acquired the supremacy over the Latins ; 
for this was not gained till the time of Tarquinius, and the very 
writers who ascribe it to Servius afterwards relate the same 
thing of Tarquinius. The foundation of the festival of the 
Ferine Latinae on the Alban mount was from very early times 
attributed to Tarquinius Priscus or Superbus, but a more cor- 
rect view entertained also by some of the ancients is, that it 
originated with the Prisci Latini. If the head of the Latins 
offered up the sacrifice there, and the Romans merely participat- 
ed in it, it was natural that in order to represent the equality 
of the two nations a counterpoise should have been formed on 
the other side, where Rome had the presidency and where the 
Latins were only guests. This was effected in the temple of 
Diana on the Aventine ; the Latins subsequently, after recover- 
ing their independence, transferred this national property to a 
grove near Aricia. In former times. Alba had been a sovereign 
city ; afterwards the Romans and Albans were united in friend- 
ship as two distinct peoples, and under Servius they joined each 
other in a federal union with a common sacrifice. This confe- 
deracy existed not only between the Romans and Latins, but also 
with the Sabines, and formed a great state, of which Rome was 
the centre, and there is no doubt that a portion of Etruria also was 
subject to it. This league we regard as the work of Servius, a 
view which recommendsitself by its simplicity and removes the 
above-mentioned contradiction. At the time when the plebe- 
ians became citizens, the Latins approached the Romans more 
closely, and stepped into the position which the plebeians had just 
quitted : so long as there existed any life in the Roman people, we 
find a constant advance of those elements which had been added 
to it, and as soon as an old element decayed, the nearest succeed- 
ed to its place; those who were first allied were first admitted 
into the state and formed into plebeian tribes. In this manner 
the whole of the Roman constitution was in the perpetual en- 
joyment of a renewed vitality, never stopping in its develop- 
ment. The Roman people ever refi'eshed and renewed itself, and 
Rome is the only state, which down to the fifth century con- 
stantly returned to its ovra principles, so that its life was ever 
becoming more glorious and vigorous, a feature which Montes- 
quieu regards as the only true movement in the life of states. 


At a later period checks were employed to repress that which 
was coming into existence, and then life began to withdraw and 
symptoms of decay became visible. Traces of this state of 
things appeared even a hundred years before the time of 
the Gracchi ; in their age it broke out and continued to increase 
for forty years, until it produced the war of the allies and 
that between Sulla and Marius, from which the people came 
forth as a disorderly multitude, which could no longer exist 
in republican unity, but necessarily required the absolute 
authority of a ruler. It is not difficult to say how Kome 
might have renewed and preserved herself for a few centuries 
longer: the road to happiness lay open, but selfish and 
foolish prejudices blinded the Komans, and when they were 
willing to strike into the right path it was too late. 

Respecting the gradual extension of the city, the most differ- 
ent opinions are current, which in the common works on 
Roman topography, such as that of Nardini, form the greatest 
chaos. Order, however, may be introduced into it. We must 
take into consideration that the form of these statements is not 
the same in all writers; for one account says that under this 
or that king a particular hill was built upon, another that it 
was included in the city, and a third again ihat the inhabitants 
of the hill were admitted to the franchise. The result of my 
investigations is as follows: The ancient city of Rome was 
situated on the Palatine ; thepomoerium of Romulus mentioned 
by Tacitus ran from the Forum Boarium across the Circus as 
far as the Septizonium, S. Gregorio, the arch of Constantine, 
the thermae of Titus, and thence back through the via sacra 
past the temple of Venus and Roma ; this whole circumference 
formed the suburb around the ancient city, and was not 
enclosed by a wall but by a mound and a ditch. At that time 
there existed on the Quirinal and the Tarpeian rock a Sabine 
town, which likewise had its pomoerium; between the two 
mounds and ditches ran the via sacra^ in which stood the 
Janus Quirini or Bifrons, a gateway on one side facing the 
Roman and on the other the Sabine town ; in times of peace 
it was closed, because then intercourse between the two towns 
was not desired; but in times of war it was opened, because 
the cities were allied and obliged to assist each other. An 
instance perfectly analogous to this exists in the Gaetulian town 
of Ghadames beyond Tripoli?, which is inhabited by two hostile 


tribes; it is divided by a wall into two parts, connected by a 
gate in the wall, which is closed during peace and opened 
during war.^ The Caelian hill was included in the city according 
to some by Romulus, according to others by Tullius Hostilius, 
and according to others again by Ancus Marcius; but the fact 
is, that the hill, which had been inhabited before, was under 
Ancus united with the city by means of a ditch, the fossa 
Quiritium, running from the ancient ditch of the pomoeriimi 
as far as the porta Capena; this ditch, the first extension of 
Rome, was made partly for draining off the water, partly for 
the purpose of protection. The soil there contains too much 
water to favour excavations, otherwise the most beautiful anti- 
quities would be found in the Circus : the obelisk was dug out 
thence in the sixteenth century. The aqua Marrana is not the 
aqua damnata of Agrlppa : in the ancient Circus there was a 
canal which drew off the water. It is there that we have to 
seek the septem viarum ricus, where Ancus made the ditch, 
perhaps as far as the sewers (^cloacae). On the Esquiline like- 
wise there was a suburb. But the Roman and Sabine towns 
were as yet separated by the Forum, which was then a swamp. 
The whole district of the Velabrum was still part of the river 
or a lake, and until it was drained, a topical union of the two 
towns was impossible. The Janus was the only road, and prob- 
ably formed a dike. 

The works ascribed to Tarquinius Priscus, the immense 
sewers or cloacae, consisting of one main arm and several 
branches, were executed for the purpose of effecting this union 
of the two towns. The main arm (^cloaca maxima), of very 
ancient architecture, is still to be seen, and still conveys the 
water into the river: its innermost vault is a semicircle, eisfh- 
teen palms ^ in width, and is enclosed in two other stone vaults 
of pejjerino (a volcanic stone from the neighbourhood of Gabii 
and Alba), one above the other, in the form of semicircles. 
The hewn blocks are all 7 3 palms long and 4 J high; they are 
fixed together without cement, and are kept in their places by 
the exactness with which they fit to one another in forming 

' This fact is related by Lyon, Narrative of Travels in Northern Africa, 
London, 1821, 4to. p. 162. The two tribes inhabiting the town are the Beni 
Walid and Beni Wasiil; but according to Lyon's account, the gate in the wall 
is closed in time of war. — Ed. 

' A Roman palm is about nine inches. — N. 

VOL. 1. H 


the vault. In the course of 2000 years, the whole structure 
has not sustained the trace of a change, aiid earthquakes, which 
destroyed the city and upset obelisks, have left it unshaken ; 
so that we may assert that it will last till the end of the world. 
This is the work which rendered it possible to give to Rome its 
subsequent and final limits : the whole quay is built of the same 
kind of stones, and shews the same architecture. 

The other sewers begin between the Quirinal and Yiminal, 
and run along under the Forum Augustum, the Forum Roma- 
num, and the Forum Boarium to the Velabrum and cloaca 
maxima ; they are equally well preserved, but lie deep below 
the surface of the earth. They were discovered in the time of 
pope Benedict XIV. They are executed in the same gigantic 
style, but of travertino, from which it is evident that they be- 
long to a later period, though probably to the time of the 
republic, perhaps to the first half of the fifth century of Rome, 
shortly before the Hannibalian war. The whole district down 
to the river, and on the other side of the Capitoline hill, was 
now inhabitable; but greater designs for extending the city 
were soon formed. It was desirable to form a high and dry 
plain possessing the advantage of not being inundated, and to 
which in times of war the country people might take refuge, 
on the north side of the Esquiline : for this purpose Servius 
Tullius constructed his great mound from the poj'ta Collina to 
the Esqviiline gate, neai'ly a mile in length, and a ditch of one 
hundred feet in breadth and thirty in depth. The soil taken 
out of this ditch formed the mound, which was lined with a 
wall on the side of the ditch and was provided with towers. 
Scarcely anything is left of this enormous work, wliich amazed 
Pliny who saw it in a state of perfect preservation, but its di- 
rection is still perceptible. In the times of Augustus and 
Pliny, when it was still perfect, it served as a public walk for 
the Romans; and Dionysius must have seen and walked upon 
it often enough. Rome now encompassed all its seven hills, 
as by this mound the Yiminal was first inclosed within the city, 
which thus acquired a circumference of more than five miles. 
Here then we have another proof of the absurdity of the opi- 
nion of Florus and others, who regarded the time of the kings 
as a period of infancy {infans in cunis vagiens) ; on the contrary, 
after the period of the kings, the greatness of Rome was for a 
\or\ii time on the decline. 



The question now is, Who was Servius Tullius? I will not 
trouble you with the story in Livy; the miracles there related 
belong to poetry and to the lay of Tarquinius, but attention 
must be paid to the Etruscan tradition about Servius Tullius 
and to the fragment of the speech of Claudius on the tables of 
Lyons, containing the account of Caeles Yibenna and Mastarna, 
from ancient Etruscan historians.^ Xot the slightest notice has 
been taken of these tables since their discovery in 1560, and 
my attention was not di'awn to them till I had published the 
first volume of my history, when I was censured by a cele- 
brated reviewer for having overlooked those documents. I 
never was so much surprised by any literary discovery, for I 
then still believed in the Etruscan origin of Eome, and thought 
that this document might diffuse an entirely new light over 
the history of Kome. Caeles Vibenna must be an historical 
personage ; he is too frecjuently and too distinctly mentioned 
to be fabidous, and his Etruscan name cannot have been in- 
vented by the Romans, as the Etruscan language was to them 
as foreign as Celtic is to us. Xor can it be doubted that he had 
a friend of the name of Mastarna. But when I examine the 
legislation ascribed to Servius Tullius, — allowing for whatever 
deductions must be made from historical certainty, especially 
in regard to chronology, though there is not the slightest 
doubt that Servius' reign preceded that of the last king, and 
that he was overthrown by Tarquinius Superbus who is tho- 
roughly historical, — when, I say, I examine this legislation, 
I find it so peaceful and so liberal, that I cannot see how a 
condoitiere of hired mercenaries (for such were his troops) could 
have drawn up such mild laws, and have wished to change the 
monarchy into a republic. The whole civil and political legis- 
lation of Servius Tullius has a completely Latin character, and 
his relation to the Latms also suggests that the lawgiver was 
of that nation. He may have been a native of Corniculum, 
and have ascended the throne contrary to established usage; 
he may have been the offspring of a marriage of disparagement 

' Comp. above p. 67, etc. : it has there been obseiTed that the following re- 
marks belong to the year 1828, and must accordingly be regarded as the last 
results of Niebuhr's investigations into this subject. — Ed. 



and the son of one of the Liiceres by a ^voman of Corniculnm 
previously to the establishment of theconnubium, and this may 
be the foundation of the story of his descent; but he surely 
was not a foreigner nor a commander of mercenaries. I have 
not the slightest doubt as to the honesty of the emperor Clau- 
dius, nor do I undervalue the importance of the Etruscan 
works (would that we had them ! much that we do possess of 
ancient literature might be joyfully sacrificed for them), but 
we must not ascribe too high a value to them. What they 
really Avere, no one could know before A. JMai's discovery (in 
1818) of the Veronensian Scholia on the Aeneid. We there 
find quotations from two Etruscan historians, Flaccus and 
Caecina, which immensely reduce the estimate of the value of 
Etruscan books for the early times, though they might perhaps 
be invaluable for the later history of that isolated nation. It 
appears that just as the Romans misunderstood the ancient 
Latin history and substituted the Tyrrhenian in its place, so 
the Etruscans adopted the traditions of the Tyrrhenians whom 
they subdued, and represented Tarchon, who acts a promi- 
nent part in Virgil, and may have occurred in the Roman 
tradition under the name of Tarquinius Priscus, as the founder 
of their empire from Tarquinii. If Claudius actually made use 
of the ancient rolls of the Etruscans, which were written back- 
wards, and are mentioned by Lucretius, he was on slippery 
ground, and hov/ much more so, if he followed Flaccus and 
Caecina, who wrote quite uncritically. Etruscan literature is 
mostly assigned to too early a period: from the Hannibalian 
war down to the time of Sulla, Etruria under the supremacy 
of Rome enjoyed profound peace, and it is to this period of 
somewhat more than a century, that most of the literary pro- 
ductions of the Etruscans must be referred. Previously to the 
social war, literature, as Cicero says, flourished in every part 
of Italy, but all knowledge of it is lost; there can be no doubt, 
however, that historical works were composed in other parts of 
Italy as well as at Rome. Now when a person read in Etrus- 
can books of Caeles Yibenna and Mastarna, and made his 
combinations, he might with some vanity have asked himself; 
" Wliat became of this Mastarna? he must surely have been 
Servius Tullius, whose birth is buried in obscurity." In this 
manner any one might hit upon this idea ; and Claudius, owing 
to the dulness of his intellect, was the very person to believe 


such a thing. In like manner, he says of the tribuni militares 
consulari potest ate: qui seni saepe octoni crearentur, though it is 
a fact that there were always six, half of them patricians and 
half plebeians, or promiscuously, or four patricians including 
the praefectus urbi; once only we hear of eight, in which case 
the two censors were included, as Onuphrius Panvinius has 
proved.^ This may have happened in one or two other in- 
stances, but at all events Claudius committed a mistake. Our 
account of ^Mastarna therefore is apparently based upon a ver}'- 
slender authority; the Etruscan annals from which Claudius 
derived his information may have been ancient, but no one 
says that they actually were ancient. I have here dwelt so 
long upon this subject because there is an evident tendency, 
which will not cease very soon, to derive information on the 
history of Rome from that of Etruria. The discovery of the 
Etruscan language, and the consequent power of deciphering- 
inscriptions in it, might be of some assistance ; but it is hardly 
conceivable that inscriptions should furnish much light, for 
history was contained in books only. 

The unity of the lay of the Tarquins from the arrival of 
Tarquinius Priscus down to the battle of lake Eegillus cannot 
be mistaken: it is a splendid subject for an epic poet and 
would have been much more worthy of Virgil than that of the 
^neid. It is credible enough, and seems to be derived from 
ancient traditions, that Servius TuUius was almost obliged to 
have recourse to force in order to carry his legislation, that he 
formed his centuries at his own discretion and on his own re- 
sponsibility, and that they in return recognised him as king a 
second time, and confirmed his laws. In antiquity, all such 
changes were carried into effect in a similar manner. It is 
further stated, that the patricians were indignant at this legis- 
lation, although it took nothing from them, and only granted 
something to the second estate; that they made attempts to 
murder the king; and that for this reason he would not allow 
them to live on the Esquiline where his house stood, but com- 
pelled them to reside in the valley below : all this derives great 
probability as a tradition from its internal consistency. The 
real tragedy, however, is said to have originated in the king's 
own house. His two daughters, the one a pious and the other 
a wicked woman, were married to the two sons of Tarquinius 
^ Liv. V. 1. with the commentators. 


Priscus : the pious one to the younger, L. Tarquinius, a gallant 
but ambitious youth, the wicked one to the elder, Aruns. The 
latter, seeing that her husband was inclined to renounce the 
throne, offered her hand to L. Tarquinius, and murdered her 
husband ; he accepted the offer and carried out her designs. 
Tarquinius, then, it is said, formed a party among the patri- 
cians, and with them concerted the murder of Servius Tullius. 
When the king appeared in the curia, he was thrown down 
the steps, and afterwards murdered in the street by the emissa- 
ries of Tarquinivis. Tullia, after having saluted her husband 
as king, on her return home drove over the corpse of her 
father, whence the street received the name of vicns sceleratus. 
Although we are not under the sad necessity of considering 
this as an authentic account, still it may be regarded as an 
historical fact, that Servius lost his life in an insurrection of 
Tarquinius, and that the latter was supported by the whole 
body of burghers, but more especially by the Luceres, his own 
party {/actio regis, gentes minores), who therefore derived the 
greater advantage from the revolution, while the first two 
tribes felt themselves oppressed. But I am as far from believ- 
ing all the particulars that have been handed down about the 
daughters of the aged king, as I am from believing the story 
of Lady Macbeth. Our habits and manners differ so widely 
from those of southern nations, that we can form no idea of the 
possibility or impossibility of their crimes ; but even admitting 
the possibility of these accounts, historical they certainly are 
not. It may be matter of history that the reign of Tarquinius 
Superbus was brilliant but extremely oppressive, and that he 
trampled the laws of Servius under foot; but the fearful mas- 
sacres belong to the poem. Tarquinius has the misfortune to 
possess a fearfid poetical celebrity, and probably to a much 
greater extent than he deserved. He cannot have entirely 
abrogated the Servian legislation: though it is possible that he 
stopped the assemblies of the plebeian tribes, abolished their 
festivals, and did not consult them on matters of legislation and 
in the election of magistrates. For the latter there cannot in 
fact have been much occasion, since the judges for capital cases 
were elected by the patricians. We read that Tarquinius exe- 
cuted enormous architectural works such as the magnificent 
Capitoline temple, after having prepared the area for it; and 
it is possible that he compelled the plebeians to perform such 


heavy task work, that many made away with themselves, and 
that in order to prevent this, he ordered their bodies to be 
nailed ou crosses; but we must here be cautious and scrupulous, 
for the detail at any rate is uncertain, nor is every thing true 
which cannot be asserted to be impossible. I am convinced 
that Tarquinius did not abolish the Servian division into 
classes, partly because it was an advantage to him to have the 
improved military system, and partly because, from the con- 
nection he formed with Latium, we must infer the equality 
of the constitutions of the two states, so that either Servius 
Tullius gave a Latin constitution to Rome, or Tarquinius 
Superbus a Roman one to the Latins. 


Although there is not the slightest doubt of the historical 
existence of Tarquinius Superbus, and although we may form 
some conception of his revolution, still the account which we 
have of the latter is more than doubtful. But a revolution 
unquestionably did occur ; and the constitution of Servius was 
to some extent suspended for the advantage of the patricians, 
especially those of the third tribe. It is surprising however 
that, notwithstanding this, the third tribe appears after this 
revolution to occupy a position inferior to that of the two 
others. But the very fact that the interests of the first two 
tribes did not harmonise with those of the third, prepared the 
way for a popular revolution. 

The statement that he entirely abolished the Servian consti- 
tution cannot be true, because in liis reign the relation of 
Rome to Latium contimied as before. According to Livy and 
Dionysius, the Latins, with the exception of Gabii, were in- 
duced to recognise the supremacy of Rome and of Tarquinius; 
but Cicero in his work De Republica, says: Universum Latium 
bello suhegit. Of a war with the Latins, there is no trace any 
where, and it must be left uncertain whether the other writers 
omitted to mention it, or whether Cicero wrote that sentence 
carelessly and thoughtlessly. It is probable, however, that 


from tlie earliest times tliere existed irreconcileable differences 
between the poetical and historical tradition. The story of 
Turnus Herdonius has a very poetical colouring. Under 
Servius, the league with Latium had been one of recipro- 
city, but that country now entered into the condition in which 
we afterwards find the Italian allies, that is, the condition of 
an unequal alliance, by which they were bound majestatem 
populi Romani comiter colere. It would appear that on the 
accession of Tarquinius at Rome, the Latins refused to renew 
the alliance which they had concluded with his predecessor. 

In the treaty between Rome and Carthage ^ we find Rome 
in possession of all the coast, not only of the Prisci Latini, but 
as far as Terracina, which then was probably still Tyrrhenian 
and not Volscian ; its inhabitants in the Greek translation are 
called vTTTjKoot. Rome concluded the treaty for them as well 
as for herself; and it was stipulated that if the Carthaginians 
should make conquests in Latium they should be obliged to 
give them up to Rome. . This treaty is as genuine as any thing 
can be, and it is a sti'ange fancy of a man otherwise very 
estimable^, to look upon it as a forgery of Polybius. Here 
then we find Latium still dependent upon Rome, and this 
dependance is expressly attested by Livy: at the beginning 
of the republic the relation was one that had been recently 
established. Afterwards,when all the country as far as Antium 
rose against Rome, the power of the latter again appears to 
be on the decline. The Feriae Latinae were an assembly of 
all the Latin people (not merely of tlie Prisci Latini) on the 
Alban mount, where accordingly the Latin magistrates must 
necessarily have presided ; but Dionysius relates that Tarquinius 
instituted the festival, and that a bull was sacrificed, of which 
the deputies of each town received a share {carnem Latinis 
accipere). The IMilan scholiast on Cicero's speech for Plancius^ 
says that there was a different tradition ; for that some ascribed 
the festival to Tar([uinius Priscus, — this is only an interpola- 

' This document was preserved in the archives of the aediles; and Polybius, 
jis he himself says, translated it not without great difficulty into Greek, since the 
Honians themselves were scarcely able to read and understand the ancient charac- 
ters. Such a treaty had to be renewed from time to time, as was often the case 
ill antifjuity, and is still the custom in the states of North Africa. — N. 

■^ U. Becker in Dalilmann's Forschungen auf dcm Gebiete der alien Geschichte- 

' Orclii, torn. v. part ii. p.255. 


tion for Tarquinius Superbus, caused by the hatred entertained 
against the latter, just as the foundation of the Capitoline 
temple was assigned to the former, — and others to the Prisci 
Latini, that is, to the earliest times. The latter statement is 
perfectly correct, for these festivals had existed long before 
Tarquinius, and were in fact as old as the Latin nation itself. 
But the other account also has some appearance of truth : it 
arose out of a misunderstanding which may easily be excused ; 
for if Tarquinius Superbus acquired the supremacy over the 
Latins, it is natural to infer that he also became the president 
at their sacrifices, just as the ^tolians during their supremacy 
did at Delphi, whence the well-known expression in inscriptions 
lepofivq/xovovvTcov AItcoXmv. 

Now in order to be able to make the best use of Latium for 
his objects, since after all he did not quite trust the Latins, 
Tarquinius did not allow their troops to form legions by them- 
selves or to serve under their own officers. He therefore 
combined the Roman and Latin legions, and then again divided 
them into two parts. The Latins had a division similar to 
that of the Romans; for both nations had centua'ies, those of 
the latter corresponding to the thirty tribes, those of the former 
to their thirty towns. Tarquinius united one Latin and one 
Roman century into one maniple, and th.epj'imus cen turio wsiS a 
Roman officer, just as in the East Indian possessions of the 
English the officers are always Europeans. Livy confounds the 
primus centurio with the primipilus. This is the origin of the 
maniples, and is the simple meaning of what Livy relates in a 
confused manner, though it is not difficult to discover his error. 

If, however, we take the separate accounts, we feel not a little 
perplexed as to what we are to believe. Tarquinius is said to 
have founded colonies at Signia and Circeii, and to have con- 
quered Gabii by a stratagem. Against the former I have 
nothing to say; but the latter is a forgery made up of two 
stories related by Herodotus about Zopyrus and Thrasybulus of 
Miletus. The treaty with Gabii however is authentic, and 
from it we must infer that Gabii was not contained in the 
confederacy of the thirty towns, the league with which had 
been settled before. The document of the treaty with Gabii 
existed in a temple as late as the time of Horace, and was one 
of the few documents that were preserved ; Gabii accordingly 
must have concluded a regular treaty of isopoHfy. 


It may easily be believed that Sextus Tarquinius committed 
tbe outrage on Lucretia; for similar things are still of every- 
day occurrence in Turkey, and were frequently perpetrated in 
the middle ages by Italian princes down to the time of Pietro 
Luigi Farnese (in the sixteenth centvuy); in antiquity similar 
crimes are met with in oligarchies and tyrannies, as is well 
known from the history of Demetrius Poliorcetes at Athens. 
Cicero is quite right in saying that it was a misfortune that 
Sextus hit upon a woman belonging to one of the most power- 
ful famihes. I readily believe that the woman tried to avenge 
herself; but the whole of the subsequent events, by which the 
story acquired individuality, and its connection with the 
campaign against Ardea, are of no historical value. The king 
is said to have been encamped before Ardea, and to have con- 
cluded a truce for fifteen years; but Ardea was dependent upon 
Rome before that time, since it occurs among the towns on 
behalf of which Rome concluded the treaty with Carthage. All 
therefore that remains and bears the appearance of probability 
is, that Lucretia was outraged, and that her death kindled 
the spark which had long been smouldering under the ashes. 

We are in the same perplexity in regard to the person of 
Brutus. He is said to have feigned stupidity, in order to 
deceive the king; and there were several traditions as to the 
manner in which he attempted to accomplish this object. 
His mission to Delphi along with the sons of Tarquinius, 
although the mission from Agylla at an earlier period cannot 
be doubted, seems to betray a later hand, and probably the 
same as introduced the stories from Herodotus into Roman 
history. It is further said that Tarquinius, in order to render 
the dignity of tribunus celerum, the highest after that of the 
king, powerless for mischief, gave the office to Biaitus. But 
there is every reason for believing that the whole of Brutus' 
idiocy arose solely from his name. Brutus is undoubtedly an 
Oscan word connected with the same root as Bruttii ; it signifies 
"a runaway slave," a name which the insolent faction of the 
king gave to the leader of the rebels because he was a plebeian. 
How is it conceivable that a great king, such as Tarquinius 
really was, should have raised an idiot whom he might have 
put to death to the dignity of tribunus celerum, for the purpose 
of rendering it contemptible? Tarquinius was not a tyrant of 
such a kind as to be under the necessity of weakening the 


state in order to govern it; lie might have given it power and 
vigour and yet rule over it by his great personal qualities; nor 
did the Romans think differently of him, for his statue con- 
tinued to be preserved in the Capitol with those of the other 

The following question formerly occupied much of my 
attention: how could Brutus who was a plebeian be tribunus 
celeruyn, since the celeres were the patrician equites ? I think 
I have discovered the solution. Most writers speak of him as 
if he had been the only tribunus celerum^ though it is certain 
that there Avere several, as is mentioned even by Dionysius, in 
his account of the priestly offices when relating the history of 
Numa. The celeres were the equites, but the plebeians too had 
their equites : now if each of the patrician tribes had its tribunus, 
is it not natural to suppose that, among the thirty tribunes of 
the plebeians, there was one who represented the plebeian 
celeres in opposition to the patricians, the plebeians thus appear- 
ing as a fourth tribe? The magister equitem, whose office is 
regarded as a continuation of that of the tribunus celerum, was 
not necessarily a patrician ; for P. Licinius Crassus was elected 
to it. This magistrate was at the head of all the eighteen 
centuries of the equites, in which the plebeians preponderated. 
In the memorable peace between the two estates in the year 
of the city 388 the plebeians again appear in the light of a 
fourth tribe, since the three festal days, which were observed 
at Eome and corresponded with the three tribes, were increased 
by one, undoubtedly because the plebeians as a body were 
treated as equal to the patricians though in the eyes of the 
patricians not so perfectly equal as to entitle them likewise 
to have three days. My opinion therefore is, that Brutus was 
tribune of the celeres for the plebeians. 

In order to give to the revolution its necessary sanction, it 
is said that Collatinus brought with him Brutus, and Sp. 
Lucretius brought Valerius. We may positively assert that 
Sp. Lucretius belonged to the Ramnes, Valerius to the Titles*, 

* The Fasti, such as we have them, mention four Valerii as sons of Volesus, viz. 
Publius Pophcola, Marcus, Manius, and Lucius; the hist or his son Cains is men- 
tioned only as quaestor. The ancient traditions, on the other hand, knew only two, 
Publius Poplicola and Marcus with the sm-name of Maximus. Wlicrever Volesus 
occurs, he is described as a Sabine; in the annals which Dionysius followed, 
he appears as one of the companions of Tatius ; while others state that he went 
to Rome by the command of oracles, which is probably the more ancient tradi- 


Collatinus to the Luceres; and Brutus, as we have above seen, 
may be regarded as a plebeian. It is universally acknow- 
ledged by the ancients, that Valerius belonged to the Titles; 
Cicero states that he was consul with Lucretius and resigned 
to him the fasces quia minor natu erat ; but Cicero here con- 
founds gentes minores with jninor natu, the less favoured tribe 
beino- called minor, for we know from Dionysius that when the 
first two tribes were placed on an equality, the numbers of 
the third were called vecorepoL (minores). Collatinus belonged 
to the gens of the Tarquinii, and was accordingly one of the 
Luceres. Brutus was a plebeian. Cicero's belief in the 
descent of the Junii Bruti from our L. Junius Brutus is un- 
doubted, and is worth more than the denial of the writers after 
the battle of Philippi, when JM. Brutus was to be regarded as 
a ho7no imitivus, that is as an outlaw. We learn even from 
Posidonius, that the question about the origin of the Bruti was 
a subject of discussion. Those who consider him to have been 
a patrician may mention various facts in support of their opinion : 
there is no doubt that many a patrician gens continued to 
exist only in some plebeian families, and a transitio ab plebem 
frequently occurred, especially in consequence of marriages of 
disparagement: the surname in such a case is usually plebeian, 
but the retaining of so illustrious a name as Brutus would not 
be sui'p rising. However, so long as the consulship was not 
open to the plebeians, no Junius occurs among the consuls. 
In the first period of the republic we read of a tribune of the 
people called L. Brutus, who became conspicuous as the author 
of an important plcbiscitum in the trial of Coriolanus (Dionysius 
also mentions him at the time of the secessio, but this is a 
forgery). This Brutus is a real personage; but, like the whole 
narrative of Coriolanus, he belongs to a different time. 

Setting aside all the dramatic points in our narrative, we 
find that after the fall of Tarquinius four tribuni celerum, were 
in possession of the government; and thus formed a magistracy 
of four men, Sp. Lucretius being at the same time princeps 

tion. To consider the four individuals as brothers, is one of the^ common ge- 
nealogical errors; Dion Cassius calls Marcus ovt\\ & gentilis oi Publius; and the 
addition which all others give to tlie Vakn-ii, Volcsi Filiiis or Nepos, arose only 
from the ordinary desire to trace all the members of a gens to one common 
ancestral hero. — N. 


senatus, and Valerius praefedus urhi. In Livy, every thing 
happens as on the stage ; he mistakes the natural and necessary 
course of events; but in Dionysius we find some important 
traces of real history. These four men were in no way autho- 
rised to bring any resolution of their own before the assembled 
citizens, for the patricians could determine upon nothing unless 
it was preceded by a senatus-consultum {Trpo/SovXevf/^a) , as in 
all the states of Greece — a fact which is repeatedly noticed by 
Dionysius. This was the case with the curies as well as with 
the centuries. The first branch of the legislature that acquired 
the initiative was the comitia tributa ; and it is this circum- 
stance which gives to the Publilian law its extraordinary im- 
portance. As long as the senate could do nothing without a 
proposal of the consuls, and the assembly of the people nothing 
without a resolution of the senate^ so long the consuls had it in 
their power to repress ahnost every movement simply by 
obstinate silence. In the present instance, it would seem that 
the proposal for abolishing the kingly dignity was illegally 
brought before the curies by the tribuni celerum ; but Livy 
suppressed the ancient account contained in the law-books for 
the sake of his own poetical narrative. The tribuni celerum 
assembled and resolved to propose the abolition of royalty ; the 
proposal was brought before the senate by the princeps senatus ; 
the senate and the curies sanctioned it, and this is the lex 
curiata. In order now to restore the constitution of Servius, 
the resolution of the curies was brought before the centuries 
also to obtain their sanction (the order is here a matter of 
indifference) ; and this is represented as if the army at Ardea 
had sanctioned the decree. 

It is by no means certain that the consulship was instituted 
immediately after the expvilsion of the kings: it is possible 
that at first Rome was governed by the four tribuni celerum^ 
but it is also possible that the number of rulers was at once 
curtailed and reduced to two. This was certainly not an im- 
provement; but it may have been prescribed in tlie Servian 
constitution with the distinct object of placing the commonalty 
on an equality with the patricians, that one consul should be 
a patrician and the other a plebeian; and thus it happened 
that of the first consuls CoUatinus was a patrician and Brutus 
a plebeian; unless their consulship was preceded by that of 
Sp. Lucretius and Valerius Poplicola. The beginning of the 
consular Fasti is mutilated, the first part being wanting. 



The consequences of the taking of Rome by the Gauls were not 
more serious for the city itself than for its history, the sources 
of which were thereby entirely destroyed. In all such cases, 
analogy and examples give us the best insight into the state of 
things, and the chronicles of many places furnish us with in- 
stances perfectly analogous in their beginnings. In my native 
coimtry of Dithmarsch, they begin about 150 years before the 
conquest of the country, after the great change which formed 
the burghers and the peasantry into one organised whole, an 
event which is not touched upon but presupposed. In a 
similar manner, the Chronicle of Cologne begins its records 
long after the city was great and flourishing: there were in- 
deed earlier records in all the towns of the middle ages, but 
they were little valued because they were too meagre, and had 
lost all their interest because living tradition was no longer 
connected with them. The chroniclers therefore began at a 
point which followed immediately after some memorable event. 
Such also was the case at Rome : there existed a history of 
the time of the republic but not from its commencement; it 
began somewhere about the secessio, and only a few incidents 
of the earlier period were recorded, such as the peace with the 
Sabines in the first consulship of Sp. Cassius, and the war with 
the Volscians. All the other events, as I have before shcAvn, 
were restored according to numerical schemes. 

I have already observed, that when the consuls were chosen 
from the two estates, Brutus represented the plebeians as after- 
wards did Sextius Lateranus. It is very remarkable, that 
with regard to all these ancient institutions, the Licinian laws 
were really and essentially nothing else than a restoration and 
a re-enactment of those of Servius. The consuls were origrin- 
ally called praetores {arpaTqiyoC in Dionysius); and this was 
their designation until the time of the decemvirate, when their 
power was weakened, and the title of consul was substituted as 
denoting something inferior. Roman etymologists were much 
perplexed in the derivation of this word ; we compare it with 
praesul, and exsul; praesul being one who is before another, 
exsul, one who is out of the state, and consul one who is with 


another, that is collega, whence consulere, to be together for the 
purpose of deliberating ; it has nothing to do with salire. The 
ancients had no idea of etymology ; and it is curious to observe 
how completely blind they were in this respect. The being 
together of a patrician with a plebeian, however, did not last 
long. It is stated that the expulsion of the Tarquins was at 
first by no means followed by bitter hostility against them, 
although an oath had been taken never again to allow a king 
to reign at Eome; so that it might almost appear doubtful 
whether the outrage said to have been committed on Lucretia 
had actually taken place. But the ancients were often incon- 
ceivably mild vmder such circumstances ; and it is also possible 
that the influence of the royal family and of the third tribe was 
still so great, that it was necessary to grant to the Tarquins the 
right of election to the consulship instead of the hereditary 
royalty. In Greek history, too, the royal families become 
<ykviq ap^LKci.: the Codrids became archons; those who were 
elected for ten years, and, at first unquestionably, even those 
who were appointed for only one year were Codrids. But 
such an arrangement did not last long, for Collatinus was 
obliged to abdicate, and the whole gens Tarquinia was ban- 
ished. It is not impossible that at that time there existed a 
Tarquinian tribe, the recollection of which was afterwards en- 
tirely lost. It is revolting to our feelings that Collatinus, the 
husband of Lucretia, should have been exiled, and if children 
of Lucretia were alive and were obliged with Collatinus to quit 
the country, their banishment would be a startling cruelty, but 
Lucretia's marriage with Collatinus belongs only to the poem, 
neque affirmare neque refellere in animo est. She was the 
daughter of Sp. Lucretius Tricipitinus ; and this circumstance is 
much more emphatically mentioned than her marriage, the 
story of which was probably intended to palliate the fact that 
not all the Tarquins were exiled, it being necessary to explain 
why, after all that happened, a cousin of the king had been made 
consul ; and this could not be done more easily than by referring 
to him the tradition of Lucretia. 

The characteristic feature of the consulship is, that it was a 
limitation of the kingly power to one year, and was elective 
instead of hereditary; it was further stripped of all priestly 
functions, and received no re/ievo?, which Cicero calls agri lati 
uberesque ?'egii, large estates which were cultivated for the kings 


by clients. These agri were now distributed among tlie com- 
monalty in order that the restoration of royalty might become 
impossible, and that the consuls might not have the same ex- 
tensive powers as the kings. The strength of the kings con- 
sisted, as among the Franks, in their retainers. Clovis was not 
allowed to appropriate to himself any portion of the booty, and 
yet he ruled as a despot, and his successors still more so ; but 
for this power he was indebted to his comitatus alone. In the 
middle ages and until the thirteenth century, the vassal of a 
king was of less importance than a common freeman who care- 
fully preserved his independence. The clients, who cultivated 
the estates of princes, were their vassals. 

The question now is, was the consulship of such a nature that 
it was necessary to elect two patricians without any restriction, 
or was it confined to the first two tribes, the Eamnes and Titles, 
to the exclusion of the Luceres, or lastly was it a representation 
of the patricians and plebeians? No one could offer himself 
for the consvilship, for at first the senate alone had the right of 
proposing candidates. The first of the above hypotheses is 
inconceivable ; for if the first two tribes or the first two estates 
had not been represented, it would have been much more natu- 
ral to institute a triumvirate. But the idea of a triumvirate 
does not occur in Roman history till a later time, a fact which 
was entirely overlooked until I discovered the trace of it in 
Joannes Lydus, an insignificant writer who had however the 
use of excellent materials. 

Of a plebeian consulship there is no trace down to the time 
of Licinius. According to the treaty with Carthage which is 
confirmed by a passage in Pliny, Horatius was elected in the 
place of Collatiniis, whereas in the common tradition Valerius 
Poplicola is called the successor of Collatinus; thus we have 
two irreconcilable statements side by side, and we are at liberty 
to exercise our criticism here as in the kingly period. The 
events assigned to the kingly period, occupying large spaces of 
time, admitted of extension and contraction ; and it is therefore 
a natural illusion to consider as more authentic the subsequent 
period, which is coimted year by year, and in which only pri- 
vate persons appear on the stage. But the period of uncer- 
tainty extends very far down, for the poem which related these 
occurrences came down to the battle of lake Regillus. The 
story of Coriolanus formed the beginning of another separate 


poem. The Fasti present the greatest differences. Three pairs 
of consuls are wanting in Livy, if compared with Dionysius, 
during the first thirty years ; in regard to one pair, Livy seems 
to have found a gap in the Fasti, and those Fasti in which 
this gap did not exist were interpolated; in the two other pairs, 
Lartius and Herminius are only secondary personages who are 
mentioned along with the heroes. The necessity of extending 
the Fasti was felt, because they did not accord with the com- 
putation of years, and new consulships were thus forged, but 
the names were not taken at random, but from extinct families 
and heroes of secondary rank, and these names were inserted 
between the consulships of the Valerii in order to conceal their 
uninternapted succession. We may therefore also form many 
conjectures upon other subjects. We know from Dionysius 
that the Horatii belonged to the gentes minores, so that the 
place of Collatinus was again filled by one of the Luceres; I 
therefore conjecture that it was perhaps intended that alter- 
nately two and two, first, one of the Raranes and one of the 
Tities, and next, one of the Luceres and a plebeian should 
be at the head of the state. This conjecture however cannot 
be followed up any further. But if Valerius was not the col- 
league of Brutus, all that is related about him must fall to the 
ground. After the death of Brutus, Valerius Foplicola is 
said at first not to have elected a successor, ard to bave built a 
stone house on the Velia. The temple of the Penates, erron- 
eously called the temple of Romulus, was situated at the foot 
of a steep hill, the Velia; the top of it, whereon stood the 
temple of Venus and Roma, and the arch of Titus, wa? the 
summa Velia, but the temple of Romulus was infima Velia. 
As the people, that is, the sovereign burghers, murmured at 
the building of a stone house, Valerius ordered it to be pulled 
down during the night, assembled the people, that is the con- 
cilium of the curies, appeared with his lictors without the axes, 
and ordered them to lower i\ie fasces before the concio, whence 
he received the name of Foplicola. The populus here, too, is 
undoubtedly the patricians or the assembly of the burghers, 
from whom the consrd derived his power, for such homage 
paid to the plebeian assembly would have been the act of 
a demagogue, and he would then have been called Plebicola. 
This beautiful narrative can have no historical value, because, 
according to the document, Valerius cannot have been consul 

VOL. I. 1 


alone, and tradition always mentions Sp. Lucretius as his first 
colleague. The reason of his not immediately filling up the 
vacant place in the consulship, is said to have been his fear of 
being opposed by one who had equal rights. Sp. Lucretius 
occurs in some Fasti as consul in the third year instead of 
Horatius, but then comes the unfortunate interpolation; and in 
order that the father of Lucretia might not be passed over, 
his consulship is transferred from the third to the first year. 

The Valerian laws are beyond a doubt; and it is a fact that 
on the whole the Servian constitution was restored. The patri- 
cians, as Livy says, endeavoured to conciliate the plebeians; 
and Sallust too states, that after the revolution the government 
was at first carried on with just laws and with fairness, but 
that afterwards it became the very reverse. The election of 
the consuls by the centuries was preserved in the ritual books, 
and is therefore not quite certain. The statement, that the 
first law passed by the centuries was the Valerian law, by 
which the plebeians obtained the right of appeal to the com- 
monalty, looks indeed very authentic, but is not so. It is quite 
possible that the first elections were made by the curies, as was 
afterwards unquestionably the case; but this is opposed to the 
express tradition that the condition of the plebes was at first 
far more favourable than afterwards.^ 


One tradition about Tarquinius states that he went to Caere 
and thence to Tarquinii, others make him go to Veii to obtain 
the assistance of the Veientines. The emigration to Caere is 
nothing else than a disguise of the jus Caeritum ewulcmdi, for 
this jMs exulandi always existed between Rome and the isopolites; 
the jus Caeritum was especially mentioned in the ancient law- 
books, and the flight of Tarquinius was believed to have 

' The rcniaiiulcr of this Lecture consists of an account of the artiticial chro- 
nology of the early Ronum history, and has been transferred from this place to 
page 3, etc. The following Lecture and a part of the next contained the account 
of the Etniscans, which has been inserted above, p. .5.5 etc., wliich seemed to bo 
a more ai)])ro])viato ))lace. 


occasioned it. The tradition of the books is that he went to 
Caere, and that of the poem that he went to Veii and led the 
Veientines against Rome. The annalists considered both as 
insufficient, and thought it most probable that he went to 
Tarquinii, where kinsmen of his might still have been living. 
Caere, whither the king's family is said to have gone, is not 
mentioned at all as having supported them during the war. 
Cicero, who saw the ancient historv of Rome without its inter- 
polations, knows nothing of a participation of the Tarquinians 
in the Veientine war; and in his Tusculanae, he merely says 
that neither the Veientines nor the Latins were able to restore 
Tarquinius. The battle near the forest of Arsia is purely 
mythical; Brutus and Aruns both fell fighting, and the god 
Silvanus loudly proclaimed the victory after 13,000 Etruscans 
and one Roman less had fallen on the field of battle. An 
account like this can be nothing else than poetry. 

Lars or Lar^ Porsena is an heroic name like Hercules amoncj 
the Greeks, Rustam among the Persians, and Dietrich of Berne 
or Etzel in the German lays; the chief heroes of such heroic 
lays are frequently transferred into history and their names 
connected with historical events. The war of Porsena is one 
of those traditions which were most generally current among 
the Romans; and it is described as a second attempt of the 
Tarquins to recover the throne. 

The Veientine war had had no eflTect, and there is no fur- 
ther account of it after the death of Brutus. Cicero undoubt- 
edly looked upon this war of Porsena in no other light than as 
the expedition of an Etruscan conqueror ; and it is certain that 
at that time the Romans were engaged in a highly destructive 
war with the Etruscans, in which they sank as Ioav as a nation 
can sink. It was nothing but republican vanity that threw 
this immediate consequence of the revolution into the shade; 
and the same feelinir o-ave rise to the dishonest concealment of 
the Gallic conquest. The tradition must have related a great 
deal about Porsena, as we may infer from the story respecting 
his monument at Clusium, whicli Pliny very credulously de- 
scribes after Varro, who derived his account from Etruscan 
books: it is this account in particular, whicli shakes my fliith 

' Lar is an Etruscan praenomen which frequently occurs on monuments and 
probably signifies hng or goi. Slartial's quantity Porsvna is false: 'm'Vibenna, 
Caecina and other words of the same termination, tlic penult is always long. — N. 



in the aiitlienticity of those books, which, to judge from this 
example, must have been of au oriental character. That monu- 
ment is described as a wondrous structure, such as never has 
existed nor could exist, like a fairy palace in the Arabian 
Nights' Tales. Pyramids stood in a circle and their tops were 
connected by a brass ring, upon which at intervals rose other 
pyramids of immense bases, and so on through several stages ; 
forming a pyramid of pyramids, a thing which could never 
have stood but must have fallen to pieces. It is inconceivable 
how Varro and still more a practical man like Pliny could 
have believed the existence of such a monstrosity, the impossi- 
bility of which must be manifest even to a boy. That it is an 
impossibility is confirmed by the fact, that neither Varro nor 
Pliny saw any traces of the work, whereas if it had really ex- 
isted, its ruins woidd be visible at this day, like those of the 
temple of Belus at Babylon. 2 There may have been an histo- 
rical Porsena, who became mythical, like the German Siegfried, 
who has been transferred to a period quite different from the 
true one ; or on the other hand there may have been a mythical 
Porsena, who has been introduced into history; but we must 
deny the historical character of every thing that is related 
about his war, which ha? an entirely poetical appearance. To 
what extent this is the case becomes evident, if we consider the 
account in its purity and stripped of all the additions made by 
the annalists. It is a peculiarity of all such poems that they 
are irreconcilable with other historical facts. 

According to the common tradition, the Etruscans suddenly 
appeared on the Janiculum, and the Romans fled across the 
river ; the poem did not even mention the conquest of the Jani- 
culum, but the Etrviscan army at once appears on the bank of 
the Tiber, ready to pass the Subllcian bridge: there three 
Roman heroes oppose them, Horatius Codes, Sp. Lartius and 
T. Herminius, probably a personification of the three tribes. 
While the Romans were breaking down the bridge, the three 
heroes resisted the enemy, then two of them, Lartius and Her- 
minius, withdrew, and Codes, Avho belonged to the tribe of 
the Ramncs, alone withstood the foe. After this, we have the 
account that tlic Etruscans crossed the river, and that the con- 
suls drew them into an ambuscade on the Gabinian road : this 

'■^ Qnatrcmprc de Quincy once had the unfortunate idea of making an archi- 
tectural re.'^toration of this monument. — N. 


is transferred entire from the Veientine war of A.U. 275, where 
the same thing occurs; the annalists made this interpolation, 
because it seemed strange to them that the poem should men- 
tion nothino; further of the war than the defence of the hridsre. 
Livy's account is ridiculously minute. We then find Porsena 
on the Janiculum. Now how is it possible that Rome could 
have suifered from such a famine as is presupposed in the story 
of Mucins Scaevola, if the Etruscans were encamped on tliat 
one hill only? for plunderers on the Roman side of the river 
were easily warded off. Livy states that Porsena carried on 
the war alone, whereas in Dionysius he appears allied with the 
Latins under Octavius Mamilius, an evident fabrication to ren- 
der it intelligible how Rome was surrounded and suffering 
from famine. There is no mention of any hostility on the 
part of the Latins, until their great war. But, the fact is that 
the Etruscans were masters not of the Janiculum only: that 
the famine was raging furiously is acknowledged by the Romans 
themselves. In this distress Mucins Scaevola, according to the 
poem, undertook to kill the king, but by mistake he slew a 
scribe, who was clad in purple, — a mistake inconceivable in 
history, and pardonable only in a poem. Mucins then told the 
king that he was one of 300 patrician youths (one of each gens) 
who had resolved to murder him; whereupon Porsena con- 
cluded peace, reserving to himself the seven Veientine j)agi, 
and keeping a garrison on the Janiculum. 

If we go into detail and ask, whether such a person as 
Mucins Scaevola ever existed at all, we come to another question 
which has been well put by Beaufort-^: how can Mucins be 
called by Livy and Dionysius a patrician or a noble youth, when 
the Scaevolae were plebeians? It is probable that the family 
of the Mucii Scaevolae appropriated this Mucins to themselves, 
and that in the ancient poems he had no other name but Caius ; 
it is not till the seventh century that two names are mentioned, 
and afterwards Scaevola (left-handed) was added; whereas the 
family of the Scaevolae derived this name from quite a different 
circumstance, Scaevola signifying an amulet. It is impossible 
to determine how much truth there may be in the story of the 
ancient Scaevola; the account which has come down to us is 
evidently poetical. 

^ This war of Porsena and the period of Camillus are treated in an excellent 
manner by Beaufort, and that period seems to have been the centre round whiih 
the other parts of his work were grouped in subordination. — N. 


Beaufort really threw great light upon tins part of Roman 
history, by showing that the peace of Porsena was something 
very different from what the Romans represented it. Pliny 
expressly states, that by it the Romans were forbidden the use 
of iron for any other than agricultural purposes ; and that hos- 
tacjeswere given is acknowledged even in the common nanrative : 
we thus see Rome in a state of perfect subjection : arma ademta, 
obsides dati, an expression which occurs so often respecting 
subdued nations. Pliny saw the treaty (nominatim comprehen- 
sum invenimus), but where, is uncertain; a tablet probably did 
not exist, but he may have found it in Etruscan books. Ta- 
citus in speaking of the conflagration of the Capitol mentions 
in no less distinct terms the deepest humiliation of the Romans 
by Porsena, sede Jovis optimi mascimi quam nun Porsena dedita 
URBE, neque Galli capta temeeare potuissent ; and what 
deditionem \facere means, is clear from the formxila which Livy 
gives in describing the deditio of Collatia to Ancus Marcius, 
from which we see that it was a total surrender of a nation, 
comprising both the country and its inhabitants, and that it 
may be compared to the mancipatio or to the in manum conventio 
of women in the civil law. To this period of subjection we 
must also refer a statement in the Quaestiones Romanae of Plu- 
tarch, who though he was uncritical made use of good materials : 
he says that the Romans at one time paid a tithe to the Etrus- 
cans, and that they were delivered from it by Hercules. Now 
a tithe was paid when a person occupied a piece of land be- 
longing to the state {qui ptiblici juris f actus erat), and the 
deliverance by Hercules denotes their liberation by their own 
sti-ength; the payment of the tithe was the consequence of 
their having given up to the Etruscans themselves, and all that 
belonged to them {feuda oblata). A still stronger proof of the 
calamity of that time is the diminution of the Roman territory 
by one third, the thirty tribes established by Servius Tullius 
having been reduced to twenty, to which, in the year A.U. 259, 
the trihus Crustumina' was added as the twenty-first.* It was 
quite a common custom with the Romans when a state was 
compelled to submit to them, to deprive it of a third part of 
its territory ; it is therefore here also evident, since tribes cor- 
respond to regions, and since out of thirty tribes we find only 

* That this number is correct — the manuscripts of Livy liave thirty-one — has 
been shewn in the new edition of the first vohune of my Roman history. — N. 


twenty, that in consequence of its surrender to Porsena, Kome, 
about the year A.u. 260, had lost one third of its territory: of 
which fact other traces are contained in the septem pagi ayri 
Vejentium, the surrender of which has already been mentioned- 
In the history^ in order to conceal the capture of the city, 
Porsena was made the champion of the Tarquins, and thus it 
seemed as if the war had, after all, not turned out so unfortu- 
nately, since its main object, the restoration of the Tarquins, 
had not been obtained. 

It is further related that after Porsena had returned home, 
he sent his son Aruns with a part of the army to Aricia, in 
order as Livy says (this is one of the passages in which he 
intentionally shuts his eyes to the truth), to shew that his 
expedition had not been quite in vain. But at Aricia, which 
Avas a very strong place, a stop seems actually to have been 
put to the progress of Porsena, through the assistance of Cuma, 
for Cumaean traditions also spoke of it : the Romans are said 
to have behaved with great generosity towards the fugitive 
Etruscans, whereby Porsena was induced to become their 
friend, to abandon the Tarquins, and to restore the seven 
Veientine pagi; after this Porsena is not again mentioned. 
Here we evidently have an awkwardly inserted piece of poetry. 
It continued to be a custom at Rome down to a late period, 
symbolically to sell the property of King Porsena previously to 
every auction; and Livy had good sense enough to see that 
this custom was not consistent with the statement that Porsena 
and Rome had parted as friends in arms (Sopv^evot). All 
becomes clear if we suppose that, after the defeat of the Etrus- 
cans at Aricia, the Romans rose and shook off their yoke, a 
supposition which gives to the story of Cloelia also a consis- 
tent meaning; otherwise her flight with the rest of the hos- 
tages must necessarily have been injurious. The great migra- 
tion of the Etruscans is connected with the statement that 
Tyrrhenians from the Adriatic sea along with Opicans and 
other nations appeared before Cuma, though in the common 
chronology there is a mistake of from fifteen to twenty years 
at the least. These Tyrrhenians were not Etruscans, but the 
ancient inhabitants of the country, who were pressed forward 
by the advancing Etruscans and moved in the direction of 

The result of all this accordingly is, that the Romans carried 


on an unequal contest against the Etruscans and tliclr king 
Poi'sena, to whom they submitted as their master; they lost a 
tliird of their territory, and of the rest they paid a tithe; the 
Etruscan power was broken at Aricia, whereupon the Romans 
took courage and rid themselves of their masters, but without 
recovering that part of their territory which lay beyond the 
Tiber, since even as late as the time of the Decemvirs the 
Tiber was their boundary, except that probably the Janiculura 
was Roman, as may be inferred from the law respecting the 
sale o^' debtors irons Tiherim. Whether the war of Porsena 
belongs to about the year to which it is assigned, whether it 
happened two or three years after the consecration of the 
Capitol, or at a later time, is an important question, in regard 
to which Livy and Dionysius contradict each other, and are 
both opposed to all the other authorities. It is easy to perceive 
that the poem about the war was inserted by the annalists, 
since the most ancient annals did not mention it at all. In 
like manner, the lay of the Nibelungen cannot be fixed cliro- 
nologically ; and Johannes jMiiller was obliged to use violence 
in order to obtain a fixed chronological point. Such poems 
know nothing of chronology. Valerius Poplicola appears in 
the battle of lake Regillus; and this determined what place 
should be assigned to the story. It is more probable, accord- 
ing to other accounts, that tlie war took place ten years later 
than is commonly supposed, that is, shortly before the begin- 
ning of the hostilities between the patricians and plebeians. I 
infer this from the statements respecting the census which I do 
not altogether reject, though I will not venture to assert that 
they are authentic in their present form: but they are certainly 
a sign of the rise and fall of the numbers of Roman citizens. 
The person with whom these statements originated, unless 
they were very ancient^ had formed a view of Roman history 
according to which the number of citizens during the period 
in question rose from 110,000 to 150,000, and again sank to 
110,000. If this rising or falling were in harmony with 
the annals, we might say that some speculator had repre- 
sented his view in this numerical scheme; but such a person 
from vanity would never have mentioned a diminution of the 
population, for we find on the contrary that in times when the 
population is decreasing the annals mention victories and con- 
quests. For this reason, 1 believe that some account, more 


ancient than the annals, intended to shew by a numerical 
scheme how Rome and Latium by unequal wars lost a part of 
their population. No one can answer for the correctness of the 
numbers, but the statement is independent of the annals. On 
this account I refer the statement — that, between the battle of 
lake Eegillus and the insurrection of the plebes, Rome was for 
a long time deprived^of one-third of its population — to the fact 
that the war of Porsena and the reduction of the Roman terri- 
tory which was its consequence belonged to this very period; 
the reduced number of citizens nearly corresponds to the loss 
of one-third of the territory; and the circumstance that it does 
not perfectly correspond, arises perhaps from the fact that only 
the plebeians were counted, not the patricians, or that some of 
the inhabitants of the lost districts emigrated to Rome. 


In the history of Rome, as in that of most other nations, the 
same events are frequently repeated, just as after the Gallic 
conquest the Latins and their allies revolted from Rome, so 
they broke through the alliance which had been established 
under Tarquinius, as soon as Rome was humbled by the Etrus- 
can conqueror. The confederacy between the two states which 
was formed under Servius Tullius, had become a union under 
Tarquinius, as notwithstanding the obscurity which hangs 
over all the detail, is clear from the combination of the Roman 
and Latin centuries into maniples. This combination is the 
more certain, as Livy mentions it in two passages, first in his 
account of the reign of Tarquinius, and secondly in the eighth 
book, where he describes the military system. The authorities 
from which he derived his information, contained testimonies 
quite independent of one another ; and he quotes them without 
understanding them, but in such a manner that we are able to 
deduce from his statements the correct view of the annalists : 
when he wrote the second passage, he was certainly not think- 
ing of the first. The relation between the two nations may 
have been arranged in such a way, that Rome alone had the 
iinpenu7n, but the Latins received their share of the booty; or 


that the two nations had the imperium alternately. But in the 
treaty with Carthage, we see that Rome had the supremacy 
and that the Latins were in the condition of perioeci. The 
result of the war, the only events of which are the conquest 
of Crustumeria which is historical, and the battle of lake 
Eegillus which is poetical, was that the Latins from the condi- 
tion o^ perioeci rose to that of equal allies, just as at Groningen 
the surrounding districts were raised to an equality with the 
city, and in all foreign transactions appeared owly as one pro- 
vince with the city. 

Tarquinius and his family are said to have been the first 
cause of the war; and I readily believe that he was not uncon- 
nected with the movement, since his family connection with 
Mamilius Octavius at Tusculum has an historical appearance, 
but we cannot possibly class the battle of lake Regillus as it is 
related, among the events of history. It never has occurred 
to me to deny that the Romans endeavoured to restore 
their dominion by war; but it is qtute a different question, 
whether a great battle was fought near lake Regillus under the 
command of the dictator Postumius, in which the Latins 
were conquered and thrown back into their former condition. 
Nay, if we may infer the cavise from its effects, which cannot 
be done as surely in moral affairs as in physical ones, the Latins 
were not by any means defeated, for they attained — after a 
considerable time, it is true — their object, a perfectly free 
alliance with Rome. The contrary might be inferred from the 
circumstance that Postumius, who is said to have been dictator 
or consul, was surnamed Regillensis ; but the Claudii too were 
called Regillani, and names derived from districts were quite 
common among the patricians; the Sergii for instance were 
called Fidcnates; Regillensis may have been taken from the 
town of Regillus, as some surnames were derived even from 
parts of the city of Rome, as Esquilinus, Avcntinus and others. 
Gentes bearing such names stood to those places in the rela- 
tion of patrons. Names derived from victories do not occur 
till very late, and the greatest generals before Scipio Africanus, 
did not derive surnames from the places of their victories, 
as Livy himself remarks at the end of th(3 thirtieth book. 

The Romans imagined that they had gained a complete victory 
in the battle, as is clear from the story about the Dioscuri : near 
lake Regillus, where the whole district consists of a volcanic 


tufo, the mark of a horse's hoof was shown in tlie stone (just as 
on the Rosstrappe in the Harz mountain), which was believed 
to have been made by a gigantic horse of the Dioscuri, a tra- 
dition which, down to the time of Cicero] lived in the mouths 
of the people. After the battle, the Dioscuri, covered with 
blood and dust, appeared in the comitium, announced the 
victory to the people, gave their horses drink at a well, and 
disappeared. Of this battle we have no accounts except those 
in which there is an evident tendency to make it appear histo- 
rical; but the poem nevertheless cannot be mistaken. The 
descriptions of the battle in Livy and Dionysius have more 
points of agreement with each other than is usual between the 
two writers, though Dionysius's description more resembles 
a bulletin, while that of Livy is fresh and animated, like the 
Homeric description of a struggle between heroes, the masses 
being entirely thrown into the background. The cessation of 
the peace between the two states had been announced a year 
before, in order that the many connections of friendship might 
be dissolved as gently as possible, and that the women might 
return to their respective homes. Tarquinius had gone to 
Mamilius Octavius, his son-in-law, and all the Latins were 
aroused. The dictator led the Romans against an army far 
superior in numbers, and Tarquinius and his sons were in the 
enemy's army. During the contest, the chiefs of the two armies 
met: the Roman dictator fell in with Tarquinius, who being 
severely wounded retreated, while the magister equitum fought 
with Mamilius. T. Herminius and the legate M. Valerius as 
well as P.Valerius fell, the last being slain while endeavouring 
to rescue the body of M.Valerius. In the end, the Roman 
equites gained the victory by dismounting from their horses 
and fighting on foot. The consul had ofiered a reward to 
those who should storm the hostile camp; and the object was 
gained at the very first assault, in which the two gigantic 
youths distinguished themselves. 

Even the ancients were greatly perplexed about M. and 
P. Valerius, for Marcus soon after re-appears as dictator, and 
Publius had died even before the battle ; both accordingly are 
described as sons of Poplicola ; but this is an unfortunate remedy, 
since a P.Valerius as a son of Poplicola again occurs in the 
Fasti afterwards. The poem however was not concerned about 
Fasti and annals : we cannot regard the two Valcrii as sons of 


Poplicola, but as the ancient heroes Maximus and Poplicola 
themselves who here fought and fell. The legend undoubtedly 
related that Tarquinius and his sons were likewise slain, and 
the statement that the king was only wounded arose from the 
record in the annals that he died at Cuma. The introduction 
of the dictator Postumius was certainly a pure interpolation, 
and the poem undoubtedly mentioned Sp. Lartius, who could 
not be wanting here, any more than M. Valerius. The reward 
offei'ed by the dictator refers to the legend of the Dioscuri, as 
in the war against the Lucanians under Fabricius, when a 
youth earned the ladder to the Avail, and afterwards, when the 
mural crown was awarded to him, was not anywhere to be 

This battle forms the close of the lay of the Tarquins, as 
the lay of the Xibehingen ends with the death of all the heroes. 
I am as strongly convinced of this now as I was eighteen years 
ago. The earliest period of Roman history is thus terminated, 
and a new era opens upon us. There is no definite time to 
which the battle can be assigned; some suppose it to have taken 
place in a.u. 255, others in A.u. 258. Some represent Postumius 
as consul, others as dictator, a sufficient proof that the account 
is not historical, for if it were, the Fasti would at any rate 
have accurately marked such an event. It is not impo.ssible 
that peace with the Latins was restored in a.u. 259 ; and if we 
were to take this statement literally, it would confirm the 
victory of lake Regillus. It might be conceived that the Latins 
were defeated there, and submitted to the condition which 
Tarquinius had established for them; but that afterwards the 
senate, from other motives, restored to thein the constitution 
of Servius TuUius; be this as it may, peace was renewed be- 
tween the Romans and Latins before the secession of the plebes. 
For many years after the battle of lake Regillus, Livy records 
nothing about the Latins, whereas Dionysius relates a variety 
of events which however are arbitrarv inventions: even down 
to the first resolution of the people that their prisoners should 
be restored to them, we know nothing of the history of this 
period, except that under Sp. Cassius, Rome concluded a treaty 
with the Latins, in which the right of isopolity or the jus 
municipi was conced.3d to them. The idea of isopolity changed 
in the course of time, but its essential features in early times 
were these: between the Romans and Latins and between the 


Romans and Caerites there existed this arrangement, that any 
citizen of the one state who wished to settle in the other, might 
forthwith be able to exercise there the rights of a citizen. This 
was called by the Greeks laoiroXLTeia, a word which does not 
occur till the time of Philip, when people began to feel the 
want of uniting in larger communities or states. Even before 
the war, a definite relation had existed between Rome and 
Latium, in which the connubium and commercium were recog- 
nised, the citizens of one state having the full right of 
acquiring quiritarian property in the other, of carrying on any 
trade and of conducting their law-suits in person and without 
a patron : they were in fact full citizens, with the exception of 
political rights. Such a relation may exist along with equality 
between the two states as well as with the supremacy of one; 
the change which now took place was that Rome recognised 
Latium as possessing equal rights with herself. Soon after the 
Hernicans also joined the league, so that then all the three 
states appeared in foreign matters as one state. This union 
ceased after tlie Gallic war. The treaty of Sp. Cassius in a.U. 
261 is not to be regarded as a treaty of peace, but as the foun- 
dation of a legal relation ; it is inconceivable how this treaty 
could have been mistaken, as was done even by the ancients, 
when they incidentally mention it. Dionysius quotes this 
treaty in words which display undoubted authenticity: he 
himself indeed can never have seen the tables in the rostra, for 
even Cicero in his speech for Balbus speaks of them in a man- 
ner which shews that he merely remembered having seen them ; 
but many Roman authors, as Macer and others, must have 
known them, and Cincius, who lived two hundred years before 
was well acquainted with them. This, like the Swiss treaties 
may be called an eternal treaty, for it was to remain in force 
as long as heaven and earth endured. But thirty years after- 
wards it became antiquated through the influence of circum- 
stances, and at a later period it was restored only for a short 
time. It established perfect equality between the Romans 
and Latins, which even went so far as to make them take the 
supreme command of the armies alternately. Either state when 
in distress was to be supported by the other with all its powers, 
and the booty was to be divided. 

This treaty contains the key to the understanding of another 
event. It is about this time that we first meet with a dictator, 


which was properly speaking a Latin magistracy, and existed 
not only in particular towns, but might, as Cato states, rule 
over the whole Latin people. It is therefore probable that the 
Romans likewise now elected a dictator, who ruled alternately 
with a Latin one, whence the imperiwn was conferred for six 
months only. Among the Etruscans, the king of each town 
had one lictor, and the lictors of all the twelve towns, when 
they united for any common purpose, were at the disposal of 
the one common sovereign. In like manner, the twelve Latin 
and the twelve Roman lictors were given to the common dic- 
tator: the two consuls together had only twelve lictors, who 
attended upon each alternately. At that time we also find 
frequent mention of a magister populi ; it is uncertain whether 
he was from the beginning the same person as the dictator, or 
whether he was elected from Rome alone, the dictatorship pro- 
bably existing only in consequence of the connection with 
Latium. A consul might have been dictator without there 
necessarily being a 7nagister populi; but whenever there was a 
magister populi, there must necessarily have been a dictator to 
represent Rome in transactions with foreign nations; for it is 
not natural that there should have been two names for the 
same office. It is probable that for a time there was a dictator 
every year, that office being sometimes given to one of the 
consuls, and sometimes to a person especially elected. 

In the history of the period which now follows, we find 
ourselves upon real historical ground : we may henceforth speak 
with certainty of men and events, although now and then 
fables were still introduced into the Fasti. That errors did 
creep in is no more than the common lot of all human affairs, 
and we must from this point treat the history of Rome like 
every other history, and not make it the subject of shallow 
scepticism to which it has already been too much sacrificed. 
A new war broke out in which Cora and Pometia fell into the 
hands of the Auruncans: afterwards these towns are said to 
have been recovered by the Romans and Latins, a statement 
which is very problematical. At the beginning of this period 
we still meet with great discrepancies and absurdities; but of 
what consequence is it that Livy relates this war twice, or 
whether it happened in A.U. 251 or A.U. 258. We may safely 
assert that there was an Auruncan war, that Cora and Pometia 
were lost, but afterwards recovered. It is a singular thing that 


Avhen a great loss was simply marked in the ancient annals 
of the Komans, the vanity of their descendants could not 
leave it as it stood, but attempted to compensate for the 
calamity by a bold lie. The deliverance of the city by 
Camillus is the most striking, though not the only instance 
in Roman history of this propensity; and Beaufort has well 
demonstrated its fictitious character; the account is in itself 
inconceivable, and is contradicted by Polybius^ who states 
that the Gauls returned with the booty to their own country 
in consequence of an inroad of the Veneti ; I do not mean to 
say that in this case the falsifier was not one of the ancient 
bards, for Camillus was as much a subject of poetry as the 
taking of Veil. In like manner every great defeat in the Sam- 
nite wars which cannot be concealed, is followed by a victory 
which is altogether unconnected with the course of events, 
and is intended to make up for the loss. The same thing occurs 
in the wars with the Volscians and Aequians. This is a com- 
mon human weakness, which in disastrous times we ourselves 
may experience. The Italians of the fifteenth century insisted 
upon being the genuine descendants of the ancient Romans ; 
and accordingly Flavins Blondus says that Charlemagne drove 
all the Lombards and other barbarians from Italy. When the 
news of the battle of Austerlitz arrived in the north of Germany, 
it was received with the greatest consternation; but a report 
soon spread and found its way even into the newspapers, that 
the French had gained a victory in the morning indeed, but 
that in the afternoon the Austrians and a part of the Russians 
rallied and most completely defeated the French. I witnessed 
similar absurdities in 1801 at Copenhagen. The history of 
Greece and of the middle ages is remarkably free of such 

I therefore believe in the invasion of the Auruncans : when 
Rome was laid low by the Etruscans, she was forsaken not 
only by the thirty towns, whose common sanctuary was the 
temple of Ferentina, but also by the coast towns which had 
been Latin, and were recognised in the treaty with Carthage 
as being under the protection of Rome. There is little doubt 
that Antium and Terracina, like the Latin towns properly so 
called, shook off the Roman supremacy and expelled the 
colonists. Both these towns were afterwards unquestionably 
Volscian, but it is an erroneous opinion that tliey Avcre so 


originally; they forin no exception to the general Tyrrhenian 
population of the coast. In an ancient Greek ethnological 
work which was certainly not an invention of Xenagoras, but 
was derived from Italiot authorities, Antium is described as a 
town of the same stock as Rome and Ardea; and Romus, 
Antias and Ardeas are brothers. Terracina did not receive it8 
Volscian name of Anxur till afterwards. These places became 
Volscian either by conquest or by voluntarily receiving Volscian 
epoeci, because they were in want of support, or lastly by being 
oblisred after their revolt from Rome to throw themselves into 


the arms of the Volscians. 

The Volscians were an Ausonian people, and identical with 
the Aiu'uncans, so that the same war is sometimes called 
Volscian, sometimes Auruncan. They are said to have come 
from Campania, and the Auruncans in Campania are known 
to have been Ausonians, Aurunici and Ausonici being the 
same words. Cora and Pometia, two Latin colonies, are stated 
to have revolted to them ; but we cannot determine whether 
they expelled the Latin colonists, or whether the taking of 
these places was a mere conquest. It is certain however that 
the Auruncans were in possession of Cora and Pometia, and 
penetrated even into Latium, where it is not impossible that 
they may have been defeated by the Romans. 


Sallust, who in the introduction of his lost history of the 
period subsequent to the death of Sulla gave, like Thucydides, 
a brief survey of the moral and political history of his nation, 
which is preserved in St. Augustin, says that Rome was ruled 
fairly and justly only so long as there was a fear of Tarquinius; 
but that as soon as this fear was removed, the potres'^ indulged 
in every kind of tyranny and arrogance, and kept the plebes 
in servile submission by the severity of the law of debt. In 
like manner, Livy states that the plebes, who down to the 

' That is, the patricians; for all coiTcct writers use the term pafre* only of 
the patriciaus and not of the senate. — N. 


destruction of the Tarqiiins had been courted with the greatest 
care, were immediately afterwards oppressed ; that until then 
the salt wliich belonged to the publicum had been sold at a low 
price, that tolls had been abolished, and that the king's domain 
had been distributed among the plebeians, in short the (f)c\dv- 
dpcoira SUaia of Servius Tullius had been restored. Lastly, 
we must notice the ancient tradition, that Brutus completed 
the senate, qui imminutus erat, with plebeians : as he was tri- 
bunus celerum of the plebeians and afterwards plebeian consul, 
it is not at all unlikely that he admitted plebeians into the 
senate, though not such a large number as is stated. But this 
cannot have been of long duration ; plebeian senators cannot 
have continued to exist down to the decem viral legislation ; 
for Sallust, who in the speech of Macer displays an uncommon 
knowledge of the ancient constitution, says, and his statement 
is believed by St. Augustin , one of the greatest minds endowed 
with the keenest judgment, that the patricians soli in imperils 
habitabant ; whence it is probable that when things became 
quiet, they expelled the plebeians. Analogies are found in 
the histories of all countries, just because it is in accordance 
with human nature. There can be no doubt that a strong 
party of the exiled royal family had remained behind, as usually 
happens in all revolutions, or a new party may have formed 
and joined the exiles, as in the Italian towns of the middle 
ages. Whatever we may think of the battle of lake Regillus, 
and however little we may believe in the existence of a cohort 
of Roman emigrants in the army of the Latins, we may with 
confidence assume that the royal exiles were joined by a large 
number of Romans, who continued to keep up a connection 
with persons of the same party in the city, as did the (^vydhe'i 
in Greece ; and as was the case in the great rebellion in Britain, 
when the Stuarts were abroad, and the Irish catholics and the 
Scotch presbyterians, who were subdued and partly expelled 
by Cromwell, joined the ancient nobles who were scattered 
about with the royal family; the same thing took place in the 
French revolution also. As long as Tarquinius, who was per- 
sonally a great man, lived, the patricians hesitated to go to 
extremes in their innovations, though they insulted the ple- 
beians and deprived them of the impci-ia ; they may even have 
expelled them from the senate, and tliey certainly did not fill 
up with plebeians those places which became vacant by death- 

VOL. I. K 


The aristocratic cantons in Switzerland were always mild 
towards the commonalty when they were threatened by out- 
ward dangers, otherwise they were harsh and cruel ; so also, 
immediately after the English revolution of 1688 the rights of 
the dissenters were far greater than twelve or fifteen years 
later. What particular rights the plebeians may have lost 
cannot be said; it is not improbable that the Valerian law 
respecting the appeal to the tribes was formally repealed ; but 
that law had previously become a dead letter, because it could 
be maintained only by bringing a charge after the expiration 
of his office against the consvd who had acted contrary to it; 
and this was a step which the plebeian magistrates no lono-er 
dared to take. But the real oppression did not begin till the 
fear of an enemy from without was removed. 

Whether the law of debt had been altered by Servius Tvdlius, 
whether Tarquinius had abolished the Servian laws, and 
whether Valerius restored them, are questions in regard to 
which we cannot believe Dionysius unconditionally. Tar- 
quinius is said to have completely destroyed the tables on 
which the Servian law was written, in order to efiiice the 
recollection of it. This sounds very suspicious ; for if only a 
single person had taken a copy, the king's measure would have 
been of no avail ; we may however infer from this statement 
that the law was not contained in the jus Popirianum : the 
plebes would surely have restored it after the secession, if they 
had been deprived of a right so expressly granted to them It 
would therefore seem that we here have one of the plebeian 

The consequence of the law of debt was a revolution. Had 
the senate and the patricians known how to act with prudence, 
and had they divided the opposition party, a thing which is 
very easy in free states, the patricians would have been superior 
to the plebeians, not indeed in numbers but in many other 
respects; for the patricians were almost the only citizens that 
had clients, and there are many passages in Livy and Diony- 
sius, from which it is evident that during the first centuries of 
Rome the number of clients was very great, that the patricians 
distributed the domain land in many small farms among them, 
and that they had them entirely in their power. These clients 
were not contained in the tribes, but through their patrons 
they were connected with the curies; they did not hold any 


hereditary property in land except by special permission of 
their patrons, so that they were altogether dependent on the 
patricians. The plebeians, on the other hand^ consisted of 
quite different elements, Latin equites, wealthy persons, and a 
number of poor people ; they were either landed proprietors or 
free labourers. These various elements might easily have been 
separated ; those who occupied a high station^ were ambitious 
to obtain offices and influence in the state, while the common 
people were unconcerned as to whether the first among them 
could obtain the consulship or not, but were anxious about 
very different things : the patricians with their want of pat- " 
riotism and justice might thus easily have separated the mass 
from the noble plebeians; but their avarice was as great as 
their ambition, and thereby they oppressed the people doubly. 
The whole of the domain land was in their possession, and if 
they had given up to the poor small portions of it, they would 
have gained them over to their side and thus detached them 
from the rest of the plebeians: but as the patricians had ex- 
clusive possession of all the trade in money, they considered 
themselves sufficiently safe. The trade in money was un- 
doubtedly of such a kind that all banking business was carried 
on by foreigners, or freedmen under the protection of a patri- 
cian, as at Athens by Pasion, who was a metoecus, and paid 
an Athenian citizen for allowing him the use of his name." 
All money transactions at Athens were in the hands of the 
trapezitne; in Italy, during the middle ages, in those of the 
Lombards ; and in our days in those of the Jews, none of whom 
have real homes: a poor plebeian may often have tried to 
borrow money of his neighbours, but was more frequently 
obliged to go to the city and procure the money at the bankers. 
The expression persona, in legal phraseology, arose from the 
fact that a foreigner was not allowed to plead his own case in 
a court of justice; and as another was obliged to do it for him, 
he made use of a mask so to speak ; the fact that in later times 
a pereginus could act for himself, and that a praetor jjeregrinus 
was appointed for this very purpose, did not arise from the 
multitude of business but from political causes. The patricians 
themselves cannot have possessed very large sums of money; 
but foreigners who went to Rome were obliged, like the clients, 
to place themselves imdcr their protection; for which the 

' Bocckli, Pull. Ecov. of Athene, p. 480. second edit. 
K 2 


patricians of course received a coiDpensation in money. It 
may, however, have happened now and then that patricians did 
business on their own account. According to this view, then, 
their usury was, after all, not so sordid as is commonly 

The civil law for patricians was quite different from that for 
plebeians, since they had come together from different states; 
the twelve tables which laid the foundation of the political 
principles by which Rome was to be governed, also first intro- 
duced one civil law for all. Among our own ancestors, too, 
the law was not varied according to geographical position but 
accordhag to persons. The native population of Italy dowji to 
the twelfth century had the Roman law, while the Germanic 
population had the Lombardic and Salic laws; but when the 
ancient municipalities were abolished and the differeiit elements 
united, it became customary to draw up regulations binding 
upon the whole population ; the people more and more forgot 
their old peculiar institutions, and thus gradually arose the 
statutes of each of the Italian towns. The law of debt for the 
patricians was liberal, but that affecting the plebeians was 
severe; it was in force among the plebeians themselves, but 
became dangerous to them only in as far as it also existed 
between them and the patricians. As soon as there is a possi- 
bility of becoming involved in debt, the number of smaU landed 
proprietors decreases from century to century. A comparison 
of the registers of Tivoli in the fifteenth century with those 
of the present day, shews that the number of landed proprietoi's 
was then fifty times greater than at present. 

Tlie general law of debt which is found in the East, among 
the Greeks and the Northern nations, as well as among the 
Romans, was, that the person who borrowed money pledged 
liimself and his family for the debt. Plutarch, in the life of 
Solon, relates that at Athens there were nearly a thousand 
bondmen for debt, who, unless they were able to pay, were 
sold into slavery. Among the Romans this personal responsi- 
bility existed in the most rigorous form: a man might pay his 
debt by personal labour, or sell his property for a certain time, 
or, if the case was a very hard one, for life, or even sell his 
own person, whereby his children wlio were yet in their father's 
])ower, likewise, per aes ct libram, came into the mancvpium of 
the pundiaser, but on condition of tlieir being permitted to 


ransom tliemselvcs. In tliis state a person continued until he 
recoyeved himself per aes et libram. The personal imprisonment 
of insolvent debtors in our own times is a remnant of that 
a ncient law, but has become me?ningless, because a more 
humane feeling has abrogated the other part. The ancient 
Germans too might transfer their free allodia and their persons 
to another and become liis clients. 

In order to escape becoming an addictiis^ a man who borrowed 
money might sell his property for a time as security, but then 
he was bound in conscience to redeem it after a certain period. 
Fides obliged the creditor not to deprive the debtor of the 
opportunity of ransoming himself, his family, or his property; 
hence Fides was so important a goddess among the Romans, 
and without her, the severitv of the law would have ruined 
CA'-ery thing. If a person failed to pay his debt, his person 
was forfeited to his creditor, that is, he became fduciarius in 
his mancipium ; this, however, could not be done simply by 
manum injicere, but required the addiction by the praetor; the 
creditor claimed the debtor's person with these words: Hunc 
ego hominem meum esse aio ex jure Quiritium. at which declara- 
tion the five witnesses and the libripens before v>diom the con- 
tract had been concluded, had undoubtedly to be present. The 
praetor then fixed a time, and if after its expiration payment 
was not made, and the debtor was unable to prove the Uheratio 
per aes et libram, he was addicted to his creditor. The ancient 
Attic law was just the same ; but Solon abolished it and intro- 
duced in its stead the Attic law of security, from which the 
later Roman law was derived; for the equites in their import- 
ant monej' transactions endeavoured to escape from the severity 
of the law of debt, by appointing foreigners as their agents, 
who were not subject to the Roman law. Hence arose the 
laws respecting the chirographa and cenfesimoe, for at Rome 
small discount business was not done at all. The addictus was 
called nejxus because he was vexu vinctus : nexus or nexum 
originally denoted every transaction that was made in the 
presence of witnesses by traditio and by weighing the money, 
which afterwards was customary only in cases of fictitious 
purchases, whereby a certain right of property was secured to 
the creditor in case of a neglect of payment. 

A debtor might frequently pay his debt by labour, and an 
able-bodied man might employ his service very usefidly at 


times when labour fetched high wages; supposing the son of 
an old man who had pledged himself was strong, the father 
would sell him to his creditor, and when the debt was paid by 
his labour he became free again. But the interest increased 
so enormously, that it was very difficult for a poor debtor to 
get rid of his burthen; if however he worked as a nexus, he at 
least paid the interest. During the period of such labour, his 
creditor exercised over him all the authority of a master over 
his slave. The numerous class of persons who paid the debts 
of others in labour is expressly mentioned by the ancients. 

Bondage for debt, however, might also arise in another way. 
A person might become a debtor even without a contract; for 
example, by neglecting to pay a legacy, or the wages of a 
labourer engaged in his service; moi'eover, if a person com- 
mitted a ci'ime he was, according to the Roman law, obliged 
to pay to the injured party a certain compensation, obligatio ex 
delicto. All these circumstances constituted a second class of 
debtors ; and in these cases there existed adclicfio without nexus, 
as was established by the twelve tables. The praetor con- 
demned a thief to pay to the person robbed double the 
amount of the stolen property; and if this was not done within 
a fixed period, he assigned him to the injured party as a bond- 
man. In like manner, if a person asserted that another had 
purchased a thing from him without paying for it, and if the 
latter could not deny the debt (aes confessum), the creditor 
inight demand the debtor's addiction for a time {viiiculam fidei), 
whereby the other was nati;rally frightened to such a degree 
that he made every effort to pay. It is only to such cases that 
the expression vinculum fidei referred, and not to the nexum ; 
for in the former, a vindicatio might take place, and the keep- 
ing of a contract is out of the question. When a Koman stood 
in nexu, that is, when he had sold himself to another in case of 
his being insolvent, as the ]\Ierchant of Venice did to Shylock, 
he pledged his property in land, however much it might be 
burthened with debts, for the laAv of the twelve tables was nexo 
solutoque idem jus esto ; the oddictus was in quite a different 
condition, for he belonged to his creditor and had no power 
over his own person. In this manner, we may clear up the 
mystery which appears in our books, when we read that debtors 
who had sold themselves (that is ncxi) nevertheless served in 


the legions.^ Livy does not enter into the question, because 
he does not see the difficulty ; and Dionysius, who does see it, 
gets into inextricable perplexities. 

This law of debt was in a certain sense as necessary as our 
strict laws relating to bills of exchange ; but abuse is unavoid- 
able, for the wealthy are not always merciful, but harsh, and 
keep to the severity of the law. The worship of ]\Iammon 
prevailed at Rome as much as in some modern countries, and 
the severity of the actual law was very oppressive ; a further 
aggravation was its being only one-sided, for when a patrician 
Avas in difficidtv, his cousins or his clients were obliged to assist 
him, whereas plebeians were in most cases obliged to borrow 
money from the patricians. A plebeian, when given over to 
his creditor, might find himself variously circumstanced : he 
might indeed have a mild master who allowed him to ransom 
himself by his labour, but he might also have fallen into the 
hands of a heartless tyrant, Avho locked him up in his ergas- 
tulum, put him in chains, and by harsh treatment endeavoured 
to induce his relatives to come forward to liberate him. 


Such was the condition of the law about the year A.u. 260, 

when all at once a state of extraordinary general distress arose, 

such as had never existed before, but such as we meet with 

again about a hundred years later, after the Gallic calamity. 

The cause of it must be sought for in the war of Porsena, whence 

we may infer that the war belongs to a considerably later time 

than that to which it is assigned by Livy. The distress led to 

disturbances, concerning the origin of which Livy's accouut 

may be tolerably well founded. An aged captain covered with 

scars, had become the bondman of his creditor, because his house 

' Among the commentators of Livy there are ingenious and learned men, who 
have -wTitten on the condition of the tiexi ; hut all their investigations have gone 
in the wrong direction, if we except the explanation given by Donjut. But 
those who wrote after him did not profit by his teaching, but returned to the old 
errors, a>:, for instance, Drakenborcli, though he quotes Doujat: a proof how 
learned men witliout a knowledge uf the world may err in sucli things. — N. 


had been burnt down and liis property carried away; he es- 
caped from the dungeon in which he had been most cruelly 
treated by his master, and appeared in the market-]3luce 
famished, covered with rags, and disfigured with bloody stripes. 
The sight of the man produced a great commotion, and the 
plebeians generally, both those who were similarly circiim- 
stanced and those who were not, refused to obey their tyrants 
any longer. Livy's account of the manner in which the tumult 
spread further and further, and how the senate at first provoked 
the people and was afterwards frightened by them, is exqui- 
sitely beautiful, and shews a profound knowledge of human 
nature; but the detail cannot be regarded as an actual tradition, 
Imt is only an historical novel. At the very time when the 
senate and the consuls had come to the fearful conviction that 
they could not rule over the commonalty unless it was willing 
to obey, the Volscians, hearing of the discord at Rome, either 
actually advanced, or a report was spread by the patricians 
that they were on their march against Rome. But however 
this may be, the senate resolved to levy an ami}'. According 
to the original constitution, the senate alone had not the power 
to declare war, but a proposal had to be made to the curies 
which had to sanction it: according to the Servian legislation, 
the proposal had to be brought before the centuries also; but 
these things were then no longer thought of, and the annalists 
mention only the senate. The senate, then, resolved to levy 
an ariny, and as the burden of the infantry fell upon the plebs 
alone, their juniores were called up according to tribes (nominu- 
tim cilahantur); theii* answering was called nomen dare, and 
their refusing nomen abnuere. Levies were on the whole made 
in the same manner, down to the latest times of the republic. 
But when the plebeians, either in consequence of oppression or 
for other reasons, refused to serve^ they did not answer {non 
reajwndehant) ; and such a silence was the most awful thing 
that could happen. As on this occasion, the plebeians did not 
answer the call, the consuls knew not what to do; and the 
plebeians loudly shouted that they would not be so foolish as 
to shed their blood for their tyrants; the booty, they said, was 
not shared by them, but was transferred to the publicum (the 
chest of the patricians) and not into the aerarium, and that they 
were becoming more and moie impoverished, being obliged to 
pledge themselves and their families to the patricians and serve 


as boudsmen. The patricians were divided among themselves; 
and Livy relates that the minores natu among the patres were 
particularly vehement in their opposition, by which he probably 
means the minores, that is, the Luceres : young patricians can- 
not possibly be conceived as members of the senate at that 
time, for it was then a real 'yepovaria. The consuls (a.U. 259) 
belonged to different parties ; Appius Claudius represented the 
interests of the wildest oligarchs, while Servilius was mild. As 
the danger was threatening, mildness alone could lead to a 
desirable result, and all attempts to levy the army by force 
were disgracefully defeated. Servilius then caused himself to 
be empowered by the senate to act as mediator. By an 
edict he summoned all whose persons were pledged, and pro- 
mised them that they, as well as their children and relatives, 
should be safe as long as the war lasted, during which time no 
creditor should be allowed to enforce the law. Hereupon the 
plebeians flocked to the standards in large numbers. With the 
army thus formed, Servilius marched into the field and re- 
turned victorious. After the close of the war, he promised the 
army to exert all his influence with the senate to obtain the 
cancelling of the contracts of debt; but the senate granted 
nothing, the army was disbanded, Appius Claudius undertook 
the administration of justice, and without any regard to his 
colleague's promise, consigned all those who had been on the 
field of battle to their creditors or compelled them to enter 
into a nexum. The remainder of the year passed away in the 
greatest commotions. The succeeding consuls, A. Virginius 
and T. Vetusius (a.u. 260), were both moderate men, a proof 
that they were elected by the centuries; for the curies would 
have chosen the most infuriated oligarchs. The senate re- 
mained obdurate ; the consuls could produce no effect upon it 
or upon the patricians. Another attempt was made to levy an 
army, but the same diflicultics presented themselves; the con- 
suls were accused of cowardice, but those who were presmnp- 
tuous enough openly to attack the plebs, were in the end 
obliged to save their fives by flight. The real danger existed 
only on market days; for the plebs were the peasantry, and 
had so much to do that they could not come to the city except 
on market days, and when they were specially summoned. 
Agriculture in Italy requires extraordinary care, for a good 
harvest cannot be expected unless the fields are weeded several 


times during the summer. The Romans plough their fields 
five, six, or seven times, and continue the Weeding until the 
corn is about three inches high. It is almost incredible how 
much labour agriculture requires in the south, though the 
produce likewise is incredible. Hence the country people 
were fully occupied the whole year round, and had no time to 
attend to matters which were not absolutely necessary; the 
only plebeians generally at Rome, were those residing in the 
city. Hence the patricians felt safe: they had among them- 
selves vigorous men, and were supported by large numbers of 
clients, so that the plebeians contained in the four city tribes 
unquestionably formed the minority, and thus it becomes intel- 
ligible how the patricians were enabled to control the plebeians 
even without a standing or mercenary army. The burghers 
in German towns likewise, as at Cologne, kept their ascendancy 
over the commonalty, although the latter was far more numer- 
ous. Such a body of oligarchs maintains itself even by its 
pride and by having many points of union. So long as the 
nature of the plebs was unknown, it must have been incon- 
ceivable that the patricians were not in greater danger, since 
if in any town the populace (for thus the plebeians are called 
in some books) rises against the wealthy, the latter are easily 

As the attempt to levy a second army failed, the question 
was: Wliat should be done? Some proposed that the promises 
of Servilius should be kept and the contracts of debt cancelled ; 
but Appius declared that the spirit of the rebels must be broken, 
and that a dictator ought to be appointed. The dictatorship 
had been instituted for the purpose of having a magistracy not 
subject to the restrictions of the consulship, and of avoiding 
not only an appeal to the curies, but also that to the tribes 
which had been introduced by Valerius. Appius wished that 
the dictator should seize and ])ut to death every one that 
refused to serve; but this senseless advice would have been 
followed by the most fearful rebellion. The foolish assembly 
indeed adopted the plan, but the good genius of Rome led the 
people to elect as dictator IMarcus Valerius^, a man distinguished 
for his mildness and kind feelings towards the plebeians. Some 

' This is tlic iiaTiic trivcn to liiin in all our authorities; and Dionysius alone 
less correctly calls liini Manias A'alerins, whicji is a mere invention, because Mar- 
cus was saifl to liave fallen in tiie liattle of lake Kegillus. — N. 


call l>iin a gentilis, and others a brother of P. Valerius Poplicola. 
He renewed the edict of Servilius; and as the Volscians, 
Aequians, and Sabines were in arms, he formed an army with- 
out any difficulty. The statement that it consisted of ten 
legions is truly ridiculous. He gave to each of the consuls a 
part of the army, reserving one for himself The Romans 
were again victorious, and on his return he demanded of the 
senate that the promises made to the people should be fulfilled; 
but the senate disregarded all promises, and declared that the 
law miTst be complied with. Valerius might now have joined 
the plebs or withdrawn: he did the latter, and resigned his 
dictatorship. One consular army, or perhaps both, were still 
under arms, and the patricians would not allow them to return, 
because as long as an army was in the field, they could exer- 
cise control over it. Dionysius expressly states that by a 
Valerian law the consuls had, by virtue of their impeyium, un- 
limited power so long as they were at a distance of one mile 
from Rome, and they could accordingly inflict military punish- 
ment upon any one who was obnoxious to them without a 
court martial. It was for this reason that the senate would not 
allow the army to return. This was a detestable policy; for 
the army could not be kept in the field for ever, and the whole 
safety of the senate depended on the conscientiousness of the 
plebs, who it was expected would not violate their military 
oath. The insurrection, however, did break out in the camp, 
though with great moderation. It is said that the soldiers at 
first intended to slay the consul, in order to be released from 
the oath which they had taken to him personally; but they 
only refused obedience, appointed L. Sicinius Bellutus their 
leader, crossed the Anio in a body, and at a distance of three 
or four miles from it encamped en a hill which was afterwards 
consecrated, and hence called Mons Sacer. The whole plebeian 
population of the city emigrated and encamped there, and 
those who remained at Rome consisted of the patricians with 
their slaves, and of the wives and children of the emigrants. 

The patricians however did not take the latter as their 
hostages; and the plebeians, on their part, abstained from all 
devastations and only foraged in the neighbourhood to satisfy 
their immediate wants. The patricians now acted a little more 
like human beings: as long as their authority was ]iot endan- 
gered, they indulged in every kind oi' effrontery and oppression, 


as we find was invariably the case down to the passing of the 
Iloi'tensian law; but as soon as their power was set at defiance, 
they became pusillanimous, and every new struggle ended in 
disgrace. They fancied the plebeians would have no courage, 
and said to one another: " This time they are sure to lay down 
their arms ; we need only assume a threatening attitude." One 
almost feels giddy at the contemplation of such madness, and 
yet it will be repeated ever and anon as long as the world lasts. 
The clahns of justice cannot be suppressed by arms; andthe 
patricians forgot that they had to deal with a noble but infuri- 
ated animal. Wlien, therefore, the plebeians planted their 
standards on the sacred mount, the eyes of the patricians were 
all at once opened. In the city the plebs possessed only two 
quarters^, the Aventine with the Vallis Murcia, and the Esqui- 
line, both very well fortified, provided with gates and un- 
questionably occupied by armed garrisons. The plebeians 
therefore might have taken Rome without difficulty, as their 
friends would have opened the gates to them ; but it would 
have been necessary to take by storm the other hills, all of 
which were fortified, as well as the Forum. If the plebeians 
had done this their country would have perished, for the sur- 
rounding nations would not have remained quiet; the conduct 
of the patres therefore appears perfectly mad, and it is incon- 
ceivable that the plebs once in arms did not proceed further. 
An explanation seems to be contained in the fact that the 
Latins were then at peace with Rome; and with the"r assist- 
ance the senate might have defied the plebeians. It is a 
remai'kable phenomenon deserving great attention, that in 
confederate republics the equality of their constitutions has no 
influence whatever upon their furnishing miitual aid, for people 
living under a democratic government frequently support the 
aristocratic government of another nation. In the great insur- 
rection of Lucerne and Berne in the year 1657, t1ie democratic 
cantons supported the oligarchic governments against the pea- 
sants. Such phenomena explain how the senate could maintain 
itself under the circumstances above described; allusions from 
the annals to this source of strength for the patricians occur in 
Dionysius, where Appius says, that the Latins would be very 

^ III the middle up;e.s, t\\{.' popolanii sis far as tlie Corsu were not gcmiiiie Romans, 
hut Shivouian.s and iVlhanese, who, under Innocent the VIII., had settkMl there, 
and continued to speak their own hmguagc as late ai? the fifteenth century. — N. 


willing to su]iport the senate against tlie commonalty, if the 
right of isopolity were granted to them. xUthough the senate 
and the patres made no use of the suggestions, yet it was 
important for them to know that should matters come to 
extremities they might have recoiu'se to such an expedient. 


According to the statement of Dionysius, the secession lasted 
four months, from August to December ; but this is merely a 
false combination based upon the fact, that the tribunes at all 
times entered upon their office on the 10 th of December. 
There was also a tradition tliat the dictator drove in the ctaviis 
on the ides of September, so that at that time there were no 
consuls at Rome. The distm"bances, moreover, were said to 
have broken out under the consuls Yu'ginius and Vetusius; 
Dionysius accordingly concluded that these consuls must have 
laid down their office at the end of August, and that the insur- 
rection lasted four months. If the office of the tribunes had 
never been interrupted, it would not be difficult to conceive 
that the time of their appointment was regulated in the same 
manner at first as afterwards; but Dionysius overlooked the 
fact that during the decemvirate, the tribuneship was abolished, 
and it is hardly conceivable that the tribunes should afterwards 
have re-entered upon their office on the same day as before, — 
they undoubtedly resumed their functions as soon as they were 
again allowed to assemble. The consuls entered upon their 
office on the 1st of August; and it seems certain that the peace 
between the two estates w-as concluded by the new consuls, 
Vetusius and Yirginius. The secession cannot have lasted 
more than about a fortnight, for the city could not have held 
out much longer, and a famine would have occurred if the 
legions had remained in possession of the fields. The rapidity 
of Livy's account also suggests only a short duration. 

I believe it is now generally acknowledged that Roman his- 
tory henceforth increases in aiithenticity ; where absurdities and 
imposbibiliiics are mixed up with it, confidence in the whole 


may indeed be shaken; but if we remove from history that 
whidi is strange and incredible, and give a clear exposition of 
the real relations of life, let no one say that thereby history is 
injured or loses in dignity: such sentiments are unhealthy and 

The patricians perceived when too late that they had gone 
too far, and were compelled to yield: in point of form they 
were obliged to submit to a great humiliation by sending 
ambassadors to the plebeians. The list of the ten ambassadors 
given by Dionysius is certainly authentic and taken from the 
libri augurales : forgeries would indeed have been carried far 
if such names were spurious. The end of the secession can 
only be understood by forming a clear notion of the state of 
affairs : we must remember that the government in the city 
could not only defend itself but could command also the allies, 
who had taken their oath to the Roman state, that is to the 
senate and populus, and looked upon the plebeians as rebels, 
so that it was by no means the numerical superiority of either 
of the two estates, which decided the question. A formal 
peace was negotiated by the feciales as between two free nations ; 
the patricians sent off ambassadors, and conducted the negotia- 
tions, notwithstanding their great humiliation, with a prudence 
in form which deserves our admiration; their object was to 
get out of the difficulties in which their mistakes had involved 
them as cheaply as possible. They could effect the reconcilia- 
tion only by strengthening themselves externally by their 
allies, or by dividing the plebeians. To do the latter, two ways 
were open to them : they might either gain over to their side 
the plebeians of distinction, whereby, however, they would 
have weakened their own power, or they might separate the 
mass of the plebeians from their leaders. The latter was the 
surest means. The debts of insolvents were cancelled, the 
addicti were declared free, and the nexum where it existed was 
dissolved, but the law of debt was not altered; an amnesty 
likewise was of course stipulated for. The cancelling of debts 
was no great loss to the creditors, since the interest paid had 
long ago exceeded the capital; fifty years later the rate of 
interest was reduced to ten per cent, but at that time it may 
have been fifty per cent. . Sully did similar things. 

The only permanent result of the secession was the establish- 
ment of the office of the tribuni plebis, whom avc :ire in the 


habit of calling tribunes of the people. This was not in reality 
an innovation : on the restoration of the tribuneship after the 
second secession, the commonalty had twenty tribunes, that is 
one for each tribe, two of whom were invested with real power. 
The tribes consisted of two decuries, and each of them had its 
president, just as in the senate there were ten decuries, each 
of which had a primus, who together formed the college of the 
decern primi. Symmetrical arrangements occur everywhere in 
ancient constitutions, whence we may deduce from a given 
fact one which is not given. When therefore we read that 
the first tribunes were two in number who elected three more, 
we may safely infer that of the actual twenty or twenty-one 
tribunes, these two were the principal, and that under the new 
circumstances they only advanced to a higher sphere of official 
activity. The difference undoubtedly was that the earlier 
tribunes had been elected each by his own tribe, just as the 
phylarchs in the Greek states were chosen each by his own 
phyle, whereas the new tribunes were elected by the whole 
commonalty. The names of the first tribimes are C. Licinius 
and L. Albinius; and Sicinius, who was the commander of the 
plebes during their secession, is mentioned as one of the three 
that were subsequently added. The plebeians who could not 
recover the rights which the Servian constitution had granted 
to them were obliged to be content with a protection against 
oppression, and their new magistracy was therefore instituted 
auxilii ferendi gratia ; the persons of the tribunes were by an 
oath declared inviolable {corpora sacrosancta), so that they 
could step in between the rulers and the oppressed and protect 
the latter. Considering the espi'it dc corps and the official 
power of the patricians, a tribune in former times would have 
had a difficult and useless task in bringing an accusation 
against a consul, since there existed another consul with equal 
powers, and both were backed by all the patricians ; the consul 
would have ordered his lictors to seize and chastise a tribune 
who dared to make an appeal against him to the commonalty. 
But whoever henceforth laid hands on a tribune was outlawed, 
and if the consul did not give effect to the declaration of out- 
lawry, the tribune might summon the consul after the expira- 
tion of his office before the court of the curies, or perhaps even 
before that of the tribes. It was formerly customary in speak- 
ing of Roman institutions not to make any distinctions between 


the (lifFcrcut periods; thus Justus Lipsius, an ingenious and 
very learned man, who as a philologer is infinitely above me, 
has by his authority done much mischief in Roman antiquities: 
whenever a magistracy or a military arrangement is mentioned, 
he and his followers speak of it as if it had always existed, and 
a tribune at the end of the third century is conceived as a 
magistrate with the same power as in the time of Cicero, as if 
he had at first possessed the same right of intercession and of 
making legislative proposals as afterwards.^ T1ie first tribunes 
can perhaps scarcely be called a magistracy of the commonalty, 
and certainly not of the state : they were in fact nothing else 
but persons in a position analogous to that of a modern ambas- 
sador, whose duty it is in a foreign state to protect the subjects 
of his own sovereign. 

Hitherto the patricians had exercised their power without 
any control, and the plebeians had no share in the administra- 
tion : hence arose the necessity for a magistracy which shoidd 
be able to afford protection against magistrates, as well as 
private individuals, whenever members of the plebeian order 
should be injured or ill treated. The house of a tribune, 
therefore, was open day and night; he was not allowed to quit 
the city, but, like a physician, was obliged to be always ready 
to give his assistance. This idea is grand and peculiarly 
Roman, for nothing analogous occurs in the whole historv of 
Greece. The tribunes moreover had the right to assemble the 
commonalty and to bring proposals before it; but at first we 
find scarcely any traces of this right having been exercised. 
The resolutions passed by the plebes on the proposal of a 
tribune were called plebiscifa, while the resolutions passed by 
the patricians were termed leges. An allusion to this occurs in 
a passage of Livy, where the Etruscans say that the Romans 
now consisted of two nations, each with its own magistrates 
and its own laws, an expression the importance of which Livy 
did not perceive. It may be said in general that Livy did not 
alter tlie materials he found, but only omitted Avhat he thought 
obscure or unnecessary. The plebiscita did not at first afl^ect 
the whole of the state; and it was not till more than twenty 
years later that they acquired the character of resolutions, 

' The siinic luis been tlic caso witli Roman topography, i'ur huildings which 
occur side \>y side on the Capitol have I)Cen regarded as works of the same 
period; hnl men of sense hke Sarti aot in a very different spirit. — N. 


which might become laAv (a. u. 283). The only real magis- 
trates of the plebeians were the nediles, a name which was also 
given to the local magistrates among the Latins; it is very 
probable tliat they acted as judges in disputes among the 
plebeians themselves, for the tribunes in the earliest times were 
not judges, though it may sometimes have happened that an 
appeal from the aediles was brought before them. In the civil 
law no change seems to have been made at that time. 

The powers of the tribunes were thus very slender and 
modest: they were partly of a negative character, and partly 
administrative in a limited way, but not at all legislative, and 
I do not believe that the tribunes had the right to propose 
any change in the civil law even for their own order: however, 
their power was a seed from wliich a tree was destined to grow 
up that was one day to overshadow all others. It is a 
singular circumstance that the election of the tribunes was 
committed to the centuries, since it would have been far more 
natural to assign it to the tribes; but this is another proof hoAV 
small were the advantages which the plebeians obtained by 
their first secession, for in the centuries the patricians exercised 
great influence through their clients, and thus about ten years 
later the patricians even succeeded in forming a party among 
the tribunes. The statement that they were elected by the 
curies is obviously false, but we may infer from it at any rate 
that they required the sanction of the curies, in order to prevent 
the election of obnoxious persons. The right of veto claimed 
by the English government on the election of Irish Catholic 
bisliops is of the same kind. According to Livy, this original 
arrangement ceased even before the Publilian law, by which 
the election was committed to the tribes, and previousl}-- to 
which Piso supposes that there existed only two tribunes. I 
believe that the number five is indeed of later origin, but I do 
not think it likely that it did not exist before the Publilian 
law ; for as this number answers to the five classes, how should 
it have been introduced at a time when tlie election no lono-er 
belonged to the classes, but to the tribes? It seems to me 
quite probable that the patricians, under the pretext of a fair 
settlement, contrived to fj^ain some advantage for themselves 
also, and in this manner I account for the otherwise inexpli- 
cable circumstance, that ten years later we find the curies 
electing the consuls instead of the centuries; it was only a 

VOL. I. L 


concession made to the plebeians that the election of one consnl 
was given to the centuries, Avhile the otiier, down to the res- 
toration of the consulship after the time of the decemvirs, 
remained in the hands of the curies. It is not impossible that 
an assignment of lands Avas made ; and it is very probable that 
a promise Avas given to restore the ancient legal relation of the 
ac/er publicus. The result of the secession was by no means as 
decided a victory of the plebeians over the patricians as our 
historians describe it: a firm basis had indeed been gained, 
and the plebeians subsequently made the best use of it, but 
the fruits could not be reaped without the greatest exertion. 

The contract between the two orders was now solemnly 
concluded, like a peace, by a sacrifice, a senatus consultum, and 
a resolution of the curies on the one hand, and of the plebeians, 
who Avere yet in arms, on the other : a curse was pronounced 
on those who should ever attempt to break the treaty, but the 
patricians did all they could to shake off the yoke. The 
deputies of the plebes and the dfcem primi of the senate offered 
up the sacrifice in common: order returned and the state of 
affairs improved, but as that which ought to have been done 
was not done, the causes of new commotions and ferments for 
a long time to come Avere left in operation. I have called this 
treaty a peace, a Avord Avhich is also used elsewhere on similar 
occasions: the Magna Charta of Liege, establishing the union 
betAveen the burghers and commonalty, was called In paix de 

The Latins were noAV rewarded for the service thev had 
done the senate, as is expressly mentioned by Dionysius on 
the authority of the excellent document Avhich forms the 
groundwork of his account: they obtained the right of isopolity 
{jus rmmicipii) in its original meaning, through the treaty of 
Sp. Cassius, which 1 haA^e already mentioned. 

These events Avhich avc see in a sufficiently clear light, 
are succeeded by the same darkness as hangs over the pre- 
ceding period, and for a time we have nothing but the 
Fasti. Livy relates the history of Coriolanus soon after the 
peace betAveen the two estates; but this cannot be its proper 
place. When a leaf of a book has been misplaced, it must 
be put right, if you do not Avish its author to talk nonsense. 

- Tlie Gcriiiaii expression for such ncovcnaiit is liichtimg. — N. 


The same is the case when an historical fact is assigned to a 
wrong time. I see no reason why I should not believe that 
during a famine at Rome a Siceliot king sent a supply of corn 
to the city; but tyrants do not appear in Sicily till some 
Olympiads after the time in which the history of Coriolanus is 
placed. I believe that Coriolanus was first impeached by the 
plebes, but no one would have dared to do this before the 
Publilian law. The Romans under Sp. Cassius could not have 
disputed about the distribution of the ager jiubUciis, if, as we 
read, the Volscians had advanced as far as Lavinium. I further 
believe that a L. Junius Brutus introduced the severe punish- 
ment for disturbing the tribunes while making their proposals, 
but he who woidd assign the history of Coriolanus to the year 
A.u. 262, could not possibly believe all these points. For this 
reason, I maintain that the story of Coriolanus does not belong- 
to this period, but to some time after the Publilian law. Cn. or 
C Marcius may perhaps have maintained himself in the Avar 
against the Antiatans, but he cannot have conquered Corioli, 
for in the same year this town belonged to the league of the 
Latin towns. The whole histor}^ must either be rejected as a 
fiction, or be assigned to qiiite a different time. But yet another 
combination has been attempted. The temple of Fortima Mulie- 
bris on the Latin road between the fourth and fifth milestone 
happened to stand on the spot where Coriolanus after his emi- 
gration was encamped and became reconciled. Xow the 
entreaties of his mother and the matrons, which may indeed 
be really historical, were connected with the name of Fortuna 
MuUehris ; and it was accordingly believed that that temple 
though the time of its foundation was known, had been erected 
in consequence of the event above referred to. But Fortuna 
Miiliebris corresponds to Fortuna Yirilis, who had her temple 
at Rome, there being a male and a female divinity like Tellus 
and Tellurao, just as the same contrast is expressed in animus 
and nnima.^ 

^ Sec above p. 80. 




LiA''T says that he should iiot wonder, if his readers were 
Avearied by his accounts of the wars with the Volscians and 
Aequians; and certainly every one must have had this feeling 
as soon as he became acquainted with Livy. The uniformity of 
these wars spoils the pleasure we have in reading the first 
decade. Wliat rendered them tedious to Livy, was the fact 
that he did not di^ade them into periods, and as, with the 
exception of what we read in Dionysius, he is our only autho- 
rity, it is difficult, and only to a certain degree possible, to 
obtain a clear view of the wars. The first period extends 
from A.u 280 to A.U. 290: the beginning is involved in great 
obscurity, and the conquests of Tarqulnius Superbus are very 
indefinite; afterwards the Volscians, under the name of Aii- 
runcans, Invaded the Latin territory; then follow a number of 
little wars till about a.u. 290, and during the latter years the 
Volscians appear in possession of Antlum, but lost it again. In 
the second period, things assumed a dlfi^erent aspect: the 
Aequians took a vigorous part in the Volscian war, Latium 
was completely crushed, and the war turned out very unfortu- 
nately for the Romans, Latins and Hernicans. This lasted till 
about A.U. 296, when the Romans concluded peace with the 
Volscians properly so called, and thus warded oif the danger. 
The terms of the peace are very remarkable. La the third 
period, the Romans continued the war against the Aequians 
alone; it was not attended with very great danger and was 
carried on languidly by both parties. There then followed a 
fresh Volscian war against the Ecetrans, who were allied with 
the Aequians. This period, being the fourth, begins with the 
great victory of A. Postumius Tubertus (a.u. 324); hence- 
forward the Romans made steady progress until the Gallic war; 
they took most of the Volscian towiis and greatly weakened 
the Aequians. Then followed the Gallic calamity, in which 
the Aequians also may have suffered severely. Afterwards 
(this is the fifth period), the wars break out anew, but with 
(juite a different character: the Aequians were then an insig- 
nificant enemy, the Volscians were in reality united with 


the Latins, and like the Latins themselves, fought for their 
independence. By dividing the wars into these five periods, they 
lose their intolerable sameness, and, at the same time, it becomes 
clear how the Volscians were enabled to maintain themselves. 

I shall not dwell upon the details of these wars, for even the 
strongest memory cannot retain them; nor are the accounts 
of them authentic, because Livy, being tired of them, read 
his authorities carelessly, and has given only a hurried descrip- 
tion. It must, however, be observed that after tlie treaty with 
the Latins, the enemy advanced in great force but made no 
important conquest until a later period ; for Circeii continued 
to be a Latin town as late as the time of Sp. Cassius. 

An event of great relative importance for Rome was the 
treaty with the Hernicans (a.u. 267). The right of isopolity 
must have existed between them even before, if it be true that 
in the reign of Tarquinius Superbus they took part in the fes- 
tival of Jupiter Latiaris; a Roman tradition mentioned them 
as allies even of Tullus Hostilius. After the Etruscan calamity, 
they must have deserted Rome like the Latins and the Tyr- 
rhenian coast towns, but the present treaty restored their old 
relations in a manner which was most advantageous for thetn. 
The Romans, Latins, and Hernicans were put on a footing of 
perfect equality, and the booty, as well as money and UukI, 
was to be divided among them in equal portions; when a 
colony was sent out, it received colonists from all the three 
people. Whether the annalists took a correct view of the 
matter (Livy and Dionysius differ very much from each other), 
or whether they merely supposed that as peace was concludet-l 
a war must have preceded, cannot be determined; but I am 
inclined to believe that tlie alliance was the result of a mutual 
want, since both nations were surrounded by the Volscians and 
Aequians, and the fortified towns of the Hernicans were of 
great importance to the Romans: a war between tlie Romans 
and Hernicans would at least have been very foolish. The 
Hernicans lived in five towns, Anagnia, Alatrvun, Ferentina, 
Frusino and Verulae, which extended from east to west and 
are remarkable for their cyclopean fortifications. According 
to statements in Servius and the Veronensian Scholiast on 
Virgil, whom Mai has edited incorrectly, the Hernicans were 
descended IVom the Marsians and Sabines; their name is said 
to have been derived from hernae which in the Sabine language 


signified rocks.^ It is strange that a nation in its own language 
should have designated itself by a mere surname, especially as 
the Marsians, INlarruciniaus and Pelignlans lived on far higher 
mountains. The Sabine origin of the Hernicans is therefore 
somewhat suspicious, but still it might be true, even though 
the derivation of their name were a mere fancy. But another 
difficulty is this: if they proceeded from the ]\Iarsians they 
must have forced their way through the Aeqidans, which is 
quite improbable; and in after-times there appears no connec- 
tion whatever between them and the Marsians. Julius Hyginus 
declares them to have been Pelasgians. 

The Hernicans were a remarkable people: they resisted the 
Romans and were respected by them on account of their bril- 
liant valour : the treaty with them is historically certain, and 
moreover that it was concluded not only with the Romans but 
also with the Latins ; whence they received a third of the booty. 
But there were nevertheless Roman antiquaries — and Dionysius 
allowed himself to be deceived by them — who imagined that 
the Romans alone had the supremacy, and hence received two- 
thirds of the booty, and the Latins the remaining third ; the 
Romans then, it is said, generously gave to the Hernicans one 
half of their share. But if the Romans and Latins together 
concluded an alliance with that brave people, it was no more 
than just that each of them should give up one-sixth of the 
booty; and Rome, according to Dionysius' own account, did 
not possess the supremacy over the Latins at all. The connec- 
tion must afterwards have been dissolved by some arrangement, 
but the fact that the Hernicans insisted on retaining their pri- 
vileges, subsequently led to their destruction. 

Sp. Cassius is far the most distinguished man of those times, 
in the obscure accounts of which the principal memorable 
events are connected with his name; first the treaty with the 
Sabines (a.u. 254), undoubtedly with isopolity to judge from 
the census lists; and next this treaty with the Hernicans. In 
the latter the relation of Rome to the Hernicans was put upon 
a footing quite different from what it had been before, just as 
the relation of Athens to her allies became altered about 01. 
100 after the battle of Naxos. When Athens founded her 
second maritime pc)\v(M-, the towns were far less dependent 
than before, and Demosthenes, in forming his great confederacy, 
acting with all the wisdom of an intelligent statesman, did not 
' .\ni(lt c(imi>nrcs llic (n'riiian Firn. — N. 

sr. CASSius. lol 

demand the supremacy for Athens, but merely that she should 
be the soul of the league. Traitors like -^schines charged him 
with degrading the dignity of Athens, because the Athenian 
deputy was not to have more influence than one from a 
Euboean town: they said that they wished to establish the 
supremacy of Athens, but they were liars. If Demosthenes 
had lived in the time of Pericles, I do not think that he would 
have acted with this spirit of moderation ; but his era was one, 
in which every thing depended upon protecting the liberty 
and independence of Greece against Philip ; hence he willingly 
concluded peace with any town that wished it, and only en- 
deavoured to direct by his intelligence and energy the pro- 
ceedings of the confederacy. Rome was placed in the same 
position by Cassius; and from this alone we must see that he 
was a great man, with a keen eye and a sound judgment. The 
Etruscan war had destroyed the Roman dominions on the 
right bank of the Tiber, the Volscians and Aequians were 
advancing, the coast towns were lost, and Rome was obliged 
to do not what she Avished, liut what she could. Later histo- 
rians, guided by a blind love for their country, wanted to deny 
such a state of things ; and Livy and the writers whom Diouy- 
sius followed, were full of absurd admiration of the greatness 
oi their ancestors, and maintained that Rome had never been 
weak. There may at that time have been fools or people like 
^schines, who declared Cassius a traitor because he regulated 
his conduct by the cii'cumstances of the case. In his third 
consulship, after the treaty with the Ilernicans, he wished to 
be just towards the plebeians also, and this leads us to speak 
of his important agrarian law. 

The nations of antiquity, in carrving on war, generally 
followed a principle of law different from that now in force. 
We regard war as a single combat between the genii of two 
states or between two imaginary states; the individual is not 
affected by it in regard to his person, liberty and property, and 
the law of war accordingly intends that he should be injured 
as little as possible, and that he should never be the immediate 
object of hostility; he is endangered only as far as it cannot be 
avoided. Among the ancients, on the other hand, hostilities 
affected every one belonging to the state; with us, the con- 
quered state indeed loses its right to rule over the country, 
while every individual continues to exist, as if no war liad 


taken place ; but the ancients entertained quite difFerent views. 
They took the whole property of the conquered and reduced 
them to a state of servitude; and this they did not only in 
wars of extermination; but even in ordinary wars the inhabi- 
tants of a conquered country lost their property: nay even 
when a place voluntarily surrendered, the inhabitants with 
their women and children came into the poAver of the con- 
queror, as we see from the formula of a deditio. In the latter 
case, the conqueror did not make the conquered his slaves, but 
they became his clients, and their landed property fell entirely 
into his hands.-. When such a place had suffered little and 
seemed to be wortli preservation, the Romans sent to it 300 
colonists, one from each gens, who were a (jypovpd or (fyvXaKrj, 
and each of whom obtained two jugera of land tor a garden; 
they further undoubtedly received the pasture land, either the 
whole or at least the greater part of it, and one third of the 
arable land, the remaining two thirds being left to the former 
inhabitants. Such was the nature of the original colonists. 
In other cases no colonists were sent out, it being thought 
unnecessary to keep a garrison in a place; and then the former 
inhabitants were sometimes driven ovit, but sometimes allowed 
to remain on condition of their paying a tax, usually a tithe. 
They then continued to live on their former property as tenants 
at will, who might be dispossessed at the pleasure of their 
masters. In those districts which had been laid waste in war 
or from which the inhabitants had been expelled, the Romans 
acted on a principle which is quite peculiar to themselves, and 
to which we find no parallel in the history of Greece. 

This principle, or tha jus agrarium, is to me the more inter- 
esting, as it was the first point that led me to a critical inves- 
tigation of Roman history; for in my earlier years I had 
occupied myself more Avith the history of Greece. When, as a 
young mail, I read Plutarch's biographies and Appian, the 
nature of the agrarian law was a perfect riddle to me. It had 
been believed that its intention was to interfere with property 
and to fix a certain limit to its extent, so that a person having 
above 500 jugera was deprived of the surplus, Avhich went to 
mcrcase the possessions of plebeians at the expense of patrician 
property. This crude notion of the law met with much favour, 
as for example, with JMachiavolli, who lived in a revolutionary 
age, and Avith Avhoni the means sanctified the end; and even 


with Montesquieu, who however looked upon a repetition of 
the past as an impossibility, since in his time every idea of 
revolution was quite foreign to men's minds. His example 
shows how bold speculative men become in matters which are 
unknown to them and appear impossible: at that time revolu- 
tionary ideas were common in an apparently quite innocent 
manner, even among men who during the revolution actually 
embraced the very opposite side. There are persons who in times 
of peace speak of their fondness for war, and revohitionary 
ideas were similarly cherished dming the profound peace of 
the eighteenth century. Such ideas, however, were dangerous 
for Europe, and when the revolution broke out, many persons 
at first found everything smooth and natural whose hearts were 
afterw^ards broken. 

As Plutarch and Appian expressly state that the law affected 
only the yi] SrjfMoaia, it was clear that something else must be 
understood by it than ordinary property. The first who ex- 
pressed an opinion that it referred to the a^er pvblicus was 
Heyne, in a program which he wrote at the time of the revo- 
lutionary confiscations; but the question, what is the affei' 
publicus remained unexamined, as in general Heyne often saw 
what was right, but rarely carried it out. The historians who 
after him wrote about the Gracchi, were quite in the dark 
respecting the agrarian law. Once, when I did not yet see my 
wav in these difficulties, I asked the great Fr. A. Wolf about 
it; but he^ with all his extraordinary intellectual powers, had 
the weakness to wish it to be believed that he knew everything, 
and accordingly not knowing what answer to give me, he 
assumed an air of not wishing to betray his secret, and said, 
" I shall one day write about it." It was by a mere accident 
that I was led to see the real nature of the ager pvblicus. It 
was at the time when servitude was abolished in Holstein: the 
peasants, both serfs and freemen, who had before transmitted 
their estates as an inheritance from father to son, were deprived 
of their possessions, and arbitrarily transferred to smaller and 
inferior estates, while their former possessions were thrown 
into large farms. These were revolting proceedings : in some 
places the peasants opposed them, but were punished in con- 
sequence, and the same was done even with estates occupied 
by freemen. My feelings were roused with the highest indig- 
nation, and the question naturally presented itself to my mind: 


" What right have they to act in this manner?" This led me 
to an investigation about leasehold property among different 
nations, and thus I came to consider the agei' publicus among 
the Romans. 

The general idea of the Italian nations was that the franchise 
was inseparable from the soil, and that all property in land 
proceeded from the state. The soil was only the substratum 
on which the pre-conceived citizenship rested. This bears a 
great resemblance to the feudal notions: for according to the 
strict feudal law there was no land at all without its feudal lord, 
all land proceeded from the sovereign as the supreme feudal lord, 
and then came the under-tenures, though practically this idea 
never existed in its full rigour. Another analogy occurs in 
the East, especially in the East Indies, where the sovereign is 
the real owner of the soil^ and the peasant possesses it only on 
precarious tenure. In the same manner, all landed property 
among the Italian nations proceeded from the state. 


When we read in Appian the statement that the ayer jjxiblicus 
was partly used for colonies and domain land, and partly let 
to farm (the latter statement is found in Plutarch only), we 
naturally ask, How is it possible that difficulties could arise? 
The Roman republic had only to make the law that no one 
should have more than a certain number of lots, and all evil 
consequences were prevented. But the fact is, that Appian 
and Plutarch misunderstood the ambiguous expression of their 
predecessor.^ I am not talking here about the letting of a 
piece of land to farm, but of a tax which was imposed on the 
estates; of corn the tenth {decuma), of fruit-bearing trees the 
fifth {quint a), and of other things in pi-oportion. Now if the 
corn was delivered in kind, the state must have built large 
store-houses; for the cattle grazing-money had to be paid, and 

' One dearly sees that this is not an invention of Appian, but an extract 
from the liistory of tlic Gracchi liy Vosidonius, who was not inferior to Polybius, 
and wliom Ap]>ian follows for tliat jicrioil, as, for the preceding one, he followed 
Dionysius, rolyhiiis, Fat)in?:, and lastly, it would seem, Rutilius. — N. 


this of course yielded a different return in different years. For 
these reasons a new system was adopted, and the produce of 
those taxes was let in farm to publicani. The forms of the 
Koman constitution have nearly always some analogy in the 
Greek states ; and this is often the case in the civil law also, but in 
the agrarian law the Eomans are quite peculiar. A Greek 
state conquered a country and founded colonies in it, but the 
possessio agri publici was not known among them, and there is 
only a single instance in which something similar occurs. 
From Xenophon's Anabasis we see that he consecrated an 
estate at Scillus to the Ephesian Artemis; the temple did not 
let this estate to farm, but received a tithe of its produce; and 
this tithe was farmed. It was not the whole produce of such 
an estate, but only a portion of it, that was given as an offer- 
ing to the deity, just as a victim was never offered as a 
oXoKavcrrov, but only a part of it was burnt in honour of the 
divinity. According to the Roman law, the state did not take 
from that which was publicum the highest possible amount of 
produce, but made known that every Romanus Quiris, who 
wished to cultivate a part of the conquered territory, might 
take it : this was the occnpatio agri publici ; the right belonged 
at first to the patricians only as the most ancient citizens, who 
mio-ht occupy a piece of land wherever they pleased. Such 
land was for the most part on the hostile frontier and in a 
state of devastation in consequence of war, whence the com- 
petition for it was not very great. There is no doubt that 
from the first the occupant was under the obligation of paying the 
decuma and quinta. It has always been overlooked that it was 
this rent which was let to farm by the government. 

The expressions agrum locare and ogrum vendere are synony- 
mous, and have the same meaning as frvctus agri vendere and 
agrum fruendum locare. A person in the possession of such 
an estate might in fact look upon it as his property as 
far as any third party was concerned, just like a leaseholder, 
from whom the owner may take the estate on certain conditions, 
but who is perfectly protected against any other party. This 
protection was afforded among the Romans by the possessoriul 
interdicts, so that the possession became heritable also. The 
state, on the other hand, might step in at any time and say, 
" I want to establish a colony here or distribute the land viritim, 
and the occupant must make room;'' to such a declaration by 


the state the occupant could make no opposition. It is, there- 
fore, clear that the state could always dispose of the ager publicus 
and declare for example, that no more than a certain number 
of jvcjera should be in the hands of an individual, because 
others would thereby be excluded, and because the excessive 
influence of one person through the immense nunjber of his 
clients, might become dangerous to the state. 

This is the great difference between property and mere 
possessio. The possessio was given by the praetor through the 
edict by which a person was called upon to take it; and the 
praetorian jus haercditatis in its origin refers to this possessio 
alone : the praetor gave possessionem bonorum secundum tabulas. 
A person might by his will bequeath his property to whomso- 
ever he pleased; but the possessio could be transferred to another 
only by sale in the presence of witnesses and by a fair contract; 
he who received it, proved his legal acceptance, and protected 
himself in his possession by the possessorial interdict; he had 
also witnesses that he had acquired the possession, neque vi 
neque clam neque precario. But Avhat was to be done when the 
possessor died? By his will he might disinherit his children 
altogether, and leave his property to the most unworthy indi- 
vidual, without the praetor in early times having power to 
interfere; but in the case of possessio, of which he was the 
exclusive source, the praetor could interfere and give his 
decision according to a principle quite different from that 
applied to property, just as the Lord Chancellor of England 
decides according to equity. Even those who, like Livy and 
Dionysius, entertain an unfair opinion of the plebes and the 
tribunes, cannot deny that the patricians were usurpers of the 
public land; and yet, according to the letter of the law, they 
might claim it, and hence it may readily be conceived that 
they appeared to be perfectly just and honest men. It is an 
important advantage gained by the study of history, that .we 
learn to judge fairly of men, and arrive at the conviction that 
honest men may belong to the most opposite parties, their 
worth being altogether irrespective of their party colours. 
This may be applied to the patricians; and when Livy and 
Dionysius, though both are anti-plebeian, say that the ager 
publicus was occu[)icd per injnrium and viro tmv dvai^SeaTiiTcov 
TraTpiKioov, they are unjust in tiieir expressions, as will be 
seen, if we go back to the original state of things. 


According to the oldest law, none but the original Ptonian 
citizens of the three ancient tribes, that is the patricians could 
be admitted to the possessio ; they received from the praetor as 
much land without any fixed limits as they thought they could 
cultivate; they paid nothing for it, and had only to employ 
their capital to render the land productive. But by the side 
of the populus, there now arose the plebs who constituted the 
real strength of Rome, formed the whole of the infantry, shed 
their blood in the wars, and made the conquests; the plebs, 
therefore, had an indisputable right to have their share in the 
conquests, which however the patricians continued to regard as 
their own exclusive property. There are distinct indications 
that even Servius TuUius had determined that no unlimited 
distributions should be made, but that one portion of the 
conquered territory should remain in the hands of the state, 
and the other be distributed among the plebeians as their real 
property. Squares were formed according to the rules of the 
science of the augurs, the lots were numbered and given to 
those who were to have shares; each lot consisted of a square 
{centuria). This is the origin of the division and assignment 
of land {assif/natio), and of the law of Servius Tullius which 
was inseparably connected with the constitution of the plebes. 
Sallust's expressions would lead us to conjecture that after the 
banishment of the kings, the Servian regulation was renewed. 
But the patricians again deprived the plebeians of this advantage, 
and it was only the a(/er regius that was distributed; afterwards 
all conquered lands remained in the hands of the patricians, 
who even exempted themselves from paying the tithe. The 
tribunes were anything but mutineers, and being the natural 
representatives of their order, they only wished to enforce its 
rights. It is not impossible that the loss of a third of the 
Roman territory in the Etruscan war fell particularly hard upon 
the plebeians. 

Sp. Cassius was the first who proposed an agrarian law, first 
to the senate, then to the curies, and at last to the centuries; 
or perhaps, first to the centuries and afterwards to the curies. 
This proposal was to re-enact the Servian law, to restore the 
decuma and quinta, to sell a portion of the conquered land, and 
to ineasure out and distribute the rest among the plebes. This 
is all we know about the Cassian law; the rest of Dionysius' 
statement shews, as, aller mature deliberation, I can confidently 


assert, the distinct marks of a writer of the second half of the 
seventh century, and is compiled with great ignorance of the 
ancient times. The senatusconsultum of which he speaks is 
utterly without foundation. The law respecting the distribu- 
tion of land is so closely connected with the whole fate of the 
plebeians, that it was probably talked of even in the negoci- 
ations for the peace on the Sacred Mount; but under Cassius 
it became a reality. There is every appearance that it was 
passed, for down to the time of the decemvirs the agrarian law 
is mentioned as a right possessed by the plebes, though they 
were not allowed the enjoyment of it. Cassius thus appears 
as a very remarkable man ; Cicero mentions him as a well- 
known person, and yet he is little spoken of. 

It is an historical fact, that in the following year, Sp. Cassius 
was executed for high treason, and that out of his property 
(ea; Cassiana familia) an offering was dedicated in the temple 
of Tellus in the Carinae. It seems probable that his execution 
by his own father was an invention made to soften down the 
glaring injustice of the deed. Even Dionysius is justly struck 
by the fact that Cassius who had then been thrice consul, 
should have been put to death by his father ; the leges annales, 
it is true, did not exist at that time, but it is nevertheless in- 
credible that a man who had been thrice consul, and had cele- 
brated a triumph should still have been in his father's power. 
Another tradition followed by Dionysius and Cicero some- 
what softens the account: the father of Sp. Cassius, it is said, 
declared in court that he considered his son guilty, and the 
latter was accordingly executed. The truth is that the quaes- 
tores parricidii summoned Cassius before the curies, and that 
the curies as his peers sentenced him to death. I'hus the 
matter becomes intelligible : he had most deeply wounded the 
members of his own order who were delighted to take ven- 
geance on him. Dionysius is puzzled by tlie account; but 
Livy avoids the diflicidty by representing Cassius as having 
been condemned by the plebes, because the tribunes were 
envious of him, — as if at that time the tribunes had had the 
power to make such proposals ! The question as to whether he 
was guilty or not was discussed by the ancients themselves : 
Dionysius considered him guilty, Dion Cassius thought him 
innocent, but God alone can know tlic truth. What he did 
was an act of the purest justice, but the same action may pro- 


ceed from the best as well as from the worst motives, and it is 
just as possible that he may have wished to promote the good 
of the state, as that he may have aimed at the kingly dignity. 
To suppose that he entertained such a thought was by no 
means so absurd twenty-five years after the banishment of the 
kings, as it was seventy years later in the case of Sp.j\laelius. 
Cassius was a very important man, otherwise he would not 
have been thrice consul, which for those times was something 
unheard of: with the exception of P. Valerius Poplicola no one 
had been so often invested with the consulship, and even in 
his case the Fasti are very uncertain. The manner in which 
Cassius concluded his treaties affords proof of a great soul ; it 
is, thei'efore, very possible that he had the purest intentions of 
wisdom and justice; for considering the spreading of the Vol- 
scians, the situation of Rome was far from being without 
danger; and it was necessary to keep all its strength together. 
A great man imquestionably he was, whether he was guilty or 
not guilty, and the faction which condemned him was detest- 
able. Dionysius has the strange statement that Cassius had 
children and that their execution also was talked of, but that 
they were spared, and that thenceforward the same mercy was 
shewn to the children of all criminals. This looks as if it were 
taken from the law books and resembles a new legal statute, 
but it may have been something quite different: we shall 
afterwards meet with a son of Sp. Cassius, and that in a place 
where we should least expect it. It is probable that the judge 
L. Cassius Longinus, a.u. 640, whose severity was almost 
cruelty, as well as the murderer of Julius Caesar, was descen 
ded from his gens: no wonder that this family attached itself 
to the plebes. The condemnation of Sp. Cassius by a Fabius, 
laid the foundation of the greatness of the Fabian family, a 
greatness to which there is no parallel in the Roman Fasti: for 
seven successive years (a.u. 269-275) one of the consuls was 
always a Fabius, just as a Valerius had been for five years at 
the beginning of the republic. The conclusion, therefore, 
naturally is that the Fabii were then in possession of supre- 
macy, and that the tribe of the Tities was represented by 



One of the disadvantages of a free government is the extra- 
ordinary difficulty of correcting any mistake that has heen 
committed ; the efforts of the government to make amends are 
rarely acknowledged by the people. An absolute prince may 
do so without weakening his authority or inciirring any 
danger ; but in a republic the case is different : if the people 
were good-natured and conscientious enough to offer the hand 
of reconciliation, things might go on well, but it is not so; 
when a government wishes to make amends to those whom it 
has offended, the first step the latter take is revenge. This 
consideration, especially if Sp. Cassius did fall a quite innocent 
victim, must serve to excuse the Roman rulers for committing 
a fresh act of violence after his death, and altering the consti- 
tution to their own advantage; for the government could not 
stop where it was, and least of all if it was conscious of a 
crime; for if they had allowed the constitution to remain un- 
changed, it was reasonable to expect that in the free election of 
the consuls by the centuries the plebeians would elect from 
among the patricians none but men like Sp. Cassius. They 
were obliged to do what Dionysius expresses so strangely in 
saying that the plebes withdrew from the elections, and that 
the noblest alone took part in them ; as if by the Servian con- 
stitution, any one except the nobles could ever have decided a 
question! The real state of the case is quite different; and I 
shall relate it as it actually occurred, reserving my proofs for 
another place. 

In the year after the death of Cassius, or even in the very 
same year, when consuls were to be elected, the election was 
not made by the centuries, but the senate nominated the can- 
didates, and the curies confirmed them But this jrave rise to 
the bitterest disputes between the plebes who were led by the 
tribunes, and the consuls; for although the tribunes at that 
time still required the sanction of the curies, yet the injustice 
was so great, that not even the mildest coidd have borne it. 
Hence the character of the tribuneships now became suddenly 
changed : up to this time there is no trace of tribunitian com- 
motions. But now the honor of their order was too much 


insulted, for on the one hand the agrarian law was not carried 
into effect ; and on the other, the government was in the hands 
of consuls who had been illegally elected. Accordingly the 
tribune Ti. Pontificius refused to allow a levy to be made, be- 
cause the people were not bound to serve under an illegal 
government: the ancient annals would hardly have preserved 
his name if his opposition had not been the first that ever pro- 
ceeded from a tribune. But an army was levied by force, the 
consuls either openly defying the tribunes and ordering the 
men who refused to answer to be seized and chastised, or caus- 
ing the houses of those who lived in the country to be set on 
fire and their cattle to be taken away, or lastly transferring the 
place where the levy was to be made from the city to the 
country, whither the tribunician power did not extend. When 
in this manner an army had been raised, the despair of the 
plebeians went so far that they would rather allow themselves 
to be butchered by the enemy, than fight for their tyrants. This 
exasperation continued for two years, and in the end rose to 
such a pitch, that the senate, as though it were a concession, 
consented that one of the consuls, should, perhaps without a 
senatusconsultum, be elected by the centuries. The consequence 
was that the consul elected by the centuries met with no oppo- 
sition on the part of the plebeians, while they resisted the other 
in every possible way. However, the times were so bad, and 
the surroundino- nations acted with such boldness towards 
Eome, that the tribunes themselves saw, that it would be better 
to put up with injustice than to allow the republic to perish. 
The plebes accordingly in the following year, A.u. 272, con- 
ceded to the senate and cuincs the election of one consul. But 
at the same time they must have acquired the right to elect 
their tribunes Avithout the sanction of the curies. Publilius 
could never have become tribune, if this change had not been 
made previously to his law. According to our traditions the 
number of tribunes must have been five, as early as that 

During this period, the Volscian'wars continued uninterrupt- 
edly, though they may not have been very important, so that 
the Latins and Hernicans alone were able to hold out against 
them. But one war weighed heavily upon Rome alone, — that 
against Veii. Veientine wars are mentioned under the kings, 
even from the time of Romulus, but they are quite apocryphal. 

VOL. I. M 


According to tlio most recent investigations, the town of Veii 
Avas about five miles in circumference, and was tluis as large 
as Rome in the time of Servius Tullius. It is very remarkable, 
that two siich large towns should have been situated so near 
each other, for the distance is not more than from twelve to 
fifteen miles; the fact shows, however, how strong was the 
contrast between the Etruscans and Latins in those times. 
Livy and Dionysius are very minute in relating the events of 
the war; and Livy believing all to be true, is very pleasing in 
his narrative. It may be regarded as authentic, that there was 
a long and difficult war against Veii. The detail in Livy 
contains nothing that is improbable; the account of the man- 
ner in which Cn. Manlius fell, and of the useless attempt to 
deceive fate especially, have an antique air. If we compare 
the accounts of this battle with those of the battle of lake 
Regillus, we shall find a considerable difference. The many 
stories about it were probably derived from the laudationes of 
the Fabian gens, which were continually repeated like the 
panegyric \6yoL iTnrdcjiioL of the Greeks. I believe that the 
plebeians always refused obedience to the consul elected by the 
patricians; the Fabii on this occasion also doubted whether 
the plebeians would obey their commands; but as the latter 
were enthusiastic in the struggle, their co-operation decided 
the issue of the battle, and the Fabii became reconciled to 
them. Through this reconciliation, everything assumed a dif- 
ferent aspect. One of the heads of the Fabii, who are called 
three brothers, but were probably gentiles, had fallen; two 
others saw that the oligarchs were throwing the republic into 
a desperate position. The Veientines were defeated, but the 
war continued ; and although the Latins and Hernicans were in 
arms, yet the Volscians spread farther and farther, and concord 
was the thing most needful. The Fabii themselves accord- 
ingly declared that the agrarian law must be conceded 
to the plebeians; and the consequence was, that none of the 
Fabii was elected patrician consul, whereas the plebeians 
chose their former friend, Kaeso Fabius for their consul. 
A most formidable commotion now arose, and the Fabii 
were looked upon by their own order as traitors; their propo- 
sals being rejected, they quitted the city in a body 306 in 
nimiber and formed a settlement on the Cremcra, being 
joined by their whole gens and some thousands of pie- 


beiaus. This must have been a settlement of a peculiar kind, 
for it was not a colony, having been formed per secessionem .- 
it was a political emigration, because the Fabii had fallen 
out with their own order; and they founded a home for 
themselves independent of Rome.^ It is said that only one of 
the Fabii survived, having been left behind at Rome as a child 
and in a state of ill health. Perizonius has sifted this account 
with great critical sagacity, and has shown how absurd it is to 
suppose that of 306 men in the prime of life all should have 
been without children, except one. The only surviving child 
moreover, appears a few years later as consul. The fact 
probably is, that the number 306, which is certainly symbolical, 
is not that of the warriors or even generals, as Livy says, but 
comprises the whole of the Fabian gens existing in the settle- 
ment, including women and children. If we were to suppose 
that they were 306 men capable of bearing arms, we should 
be obliged to estimate the number of all the patricians at an 
amount beyond all possibility. There can be no doubt that 
they had a large number of clients ; and the fact of the latter's 
emigrating with them is a remarkable instance of the relation- 
ship existing between patrons and clients. ^ 

The destruction of the Fabii on the Cremera is an historical 
fact, but the account of it is partly poetical, partly annalistic. 
The poetical story was, that the Fabii, trusting to the peace 
concluded with the Etruscans, went from the Cremera to Rome 
for the purpose of offering up a sacrum gentilicium in the city, 
— such a sacrifice indeed could be offered only at Rome, and 

' It was probably an attempt to conquer the Veientines by the establishment 
of a fortified place in their territory, like the 67riTeix"''j"oy of Decclea against 
Athens, for in those times a campaign lasted only a very short period, from a 
week to a fortnight; the garrison of a place either went out to meet the enemy, or 
shut themselves up within their own walls; and in order to present the inhabit- 
ants quietly returning to then' fields after the departure of the enemy, the latter 
often founded a fortified place in the territory which they had invaded. — N. 

^ Li^y says of the Fabii that they went out infelici via porta Camientali ; 
and Ovid, Carmentis portae dextro via proximo Jano est : Ire per hanc noli, 
quisqiies es : omen habet. This must be understood thus: all Roman gates had 
a double arch, through one of which people went out of, and tlu-ough the other 
into, the city; the former was called Janus dexter, and the latter Janus sinister. 
The Cannental gate was situated between the Capitoline and Quirinal. Now 
as any one who wanted to go out was not allowed to pass through the left 
Janus, he was obliged to take a round-abt)Ut way, if the place he wanted to go 
to was close to the Cannental gate: for the right Janus wa.s ominous, as behig 
that through which the Fabii had left the city.— N. 

M 2 


nil the members of the gens were obliged to attend — and not 
suspecting that the Etruscans had any hostile intentions, they 
proceeded without arms. But the Veientines roused their 
kinsmen and occupied the road Avhich the Fabii had to pass; 
tlie latter were surrounded by many thousands, who however 
did not venture to attack them in close combat, but killed 
them from a distance with slings and arrows. The sacrum 
gcntihcium was undoubtedly the statum sacrificium of the Fabian 
gens on the Quirinal which is mentioned in the Gallic 

The other account is, that the Fabii, being drawn away 
farther and farther by flocks feeding in the neighbourhood, 
and after at length coming into a woody plain, were slain by 
a numerous Etruscan army. The clients are not again men- 
tioned, but the fortress on the Cremera was taken bv the 
Veientines. We might be tempted to suspect treachery here, 
and that tlie rulers of Rome perfidiously delivered the fortress 
up to the enemy: one of the Roman consuls, T. Menenius, is 
said to have been in the neighbourhood, and to have after- 
wards beeii criminally accused; but that suspicion seems hardly 
probable, and if the consul acted treacherously, it can have 
been only from personal hatred. The same consul was defeated 
and fled to Rome, and the fugitives threw themselves into the 
city, and did not even maintain the Janiculum, the garrison 
of which fled with them; the other consul appeared just in 
time to ward off the greatest danger, and it was with difliculty 
that the I'ridge was broken down. It is true there was a 
wall also running from the Capitol to the Aventine, which 
protected the city on this side of the river; but ihe breaking 
down of the bridge was necessary in order to Isolate the 
suburb, which no doubt existed as early as that time. The 
Veientines were now masters of the whole plain ; they pitched 
their camp on the Janiculum, crossed the river, and plundered 
all the Roman territory on the left bank of the Tiber. It was 
then about the middle of summer, and the new consuls entered 
upon their office on the first of August. The enemy had 
crossed the river unexpectedly on rafts, and thus it may have 
happened that the greater part of the harvest was destroyed, 
the farms burnt down, and that men and cattle fell Into the 
hands of the enemy: the distress in the city rose to an extra- 

'■' Livv. V. 40. 


ordinary height The Roman armies were encamped outside 
the city, and hard pressed by the Veienlines. But despair 
gave them courage, and they resolved upon a daring enterprise, 
which Avas to decide whether Eome should perish or be saved. 
They crossed the river, defeated the Etniscans, and while one 
part stormed the Jauiculum, another made an attack from 
above; they lost indeed an immense number of men, but they 
drove off the enemy. I have already observed, that this ac- 
count presents a striking resemblance to that of the war 
with Porsena. One year later a truce was concluded with 
the Veieutines for forty years of ten months each, and was 
honestly kept. 

After tliese occurrences, the character of the tribuneship 

shews itself in a peculiar manner: the tribunes summoned the 

consuls of the preceding year before the people, not as our 

authors represent it before the plebes, for they were yet much 

too weak to sit in judgment on the sovereign magistrates, nay 

not even before the centuries, which were for the most part 

plebeian; but it was either not the tribunes at all but the 

quaestors that summoned the consuls, or what is much more 

probable, a great change had taken place by which the tribunes 

were enabled to give effect to their right of accusing the 

consuls before their own peers, that is, the popidus, because the 

magistrates who were bound to do so neglected their duty. 

After the consuls were condemned to pay a considerable fine, 

the tribunes proceeded to bring an accusation against their 

successors. They were acquitted, but the exasperation rose 

hii^her and hia-her. The tribunes had l^rouo-ht their accusation 

before the burghers, and the case was one on which they had 

the power to decide, for it was nwjestas populi Romaniimminuta 

re male gesta, and consequently a a-imen majestatis : but the 

tribunes now jn'oceeded further. They summoned all the 

consuls that had been in office since the time of Sp. C'assius, 

before the plebeian commonalty, because they had not done 

justice to the people in regard to the agrarian law; and this 

step was taken according to an old Italian maxim, that when 

two nations were united by a treaty any complaint respecting 

a violation of the treaty should be brought before the injured 

people. It is repugnant to our views that a person should be 

the judge in his own case, but the practice existed among all 

the ancient Italian nations, so that the Romans even followed 


the principle of delivering up Roman citizens to an allied 
nation which had been offended by them ; as examples, I may 
mention the surrender of IMancinus to the Numantines, of 
Postumius and his companions to the Samnites after the 
Caudine defeat, and of Fabius, who had offended the 
ambassadors of Apollonia. The surrender of those qui in noxa 
sunt was a general demand whenever there occurred a rerum 
repetitio. This principle is not found among the Greeks; it 
is based partly upon the noble idea that an oath before the 
actual trial is sufficiently binding, and partly upon a notion 
which is also found among the ancient Germans: with them 
any member of a family was obliged to come forward as a 
witness in a case affecting members of his own family, when 
he was called upon to do so (consacrainentales) ; a custom 
which rested upon the noble idea of fidelity. It was a principle 
that no one could judge a member of his own order but only 
defend him; from which however frightful abuses arose. It 
is surprising how impartial courts of justice at Rome sometimes 
were; to be so, however, was less difficult on account of the 
circumstance, that the accused, up to the moment when the 
verdict was given, was at liberty to retire from Rome and 
betake himself to some one of the many allied towns. At 
Caere, for example, a Roman might demand to be received as 
a citizen. The origin of this right of withdrawing and claim- 
ing the right of citizenship elsewhere was ti*aced in Roman 
books to the times of T. Tatius, who refused to deliver up his 
kinsmen to the inhabitants of Lavinium who had been injured 
by them : in consequence of this he was murdered, but after- 
wards tlic Romans surrendered the offenders to the Lavinians, 
and tlie latter the murderers of T. Tatius, that they might be 

It was upon this principle that the tribune Cn. Genucius, who 
belonged to a family which even at that time was great, sum- 
moned the patrician magistrates before the commonalty. He 
had promulgated his accusation against the consuls of the pre- 
ceding year in trinundinum, and the plebeians themselves were 
to judge; their right to do so was by no means doubtful, ac- 
cording to the treaty solemnly sworn to upon the Sacred 
INIount; nor was the issue of the trial uncertain. But in the 
exasperation of parties, the patricians resolved upon the quickest 
expedient — they committed the monstrous crime of murdering 
Genucius; and witli \\\\^ murder the accusation dropped. 



DiONYSiUS justly observes that if tlie assassins of Genucius 
had been satisfied with their crime, the terror which they 
created might have been sufficient for their purpose. The 
tribunes were in the greatest alarm, for their sacred right was 
violated; as it was necessary for their houses to be open day 
and night, no precaution could protect them against a similar 
outrage, nor against the intrusion of disguised assassins; and 
even the boldest dreads such a danger. The murderers of 
Genucius were not discovered, and the general terror para- 
lyzed everybody. The patricians exulted in their deed, and 
wanted to avail themselves of the first moment for making a 
levy, and for adding scorn and insult to their crime: their 
intention was to select the noblest of the plebeians, and in the 
field to put them to death or abandon them to the enemy. 
But they were too hasty in their insolence, and their exultation 
knew no patience : they summoned a distinguished plebeian, 
Volero PubliHus, who had before been centurion, and wanted 
to enlist him as a common soldier. Distinguished and wealthy 
families existed among the plebeians as well as among the 
patricians ; and to these the Publilii belonged. Wlien Publilius 
refused to obey, the consuls sent their lictors to drag him 
obtorto C0//0 before their tribunal, to strip him, and scourge him 
servili modo. The Roman toga was a very wide garment of 
one piece in the form of a semicircle ; there was no seam in it, 
and a man might wrap himself entirely up in it : now if a 
person was to be led before a magistrate, the lictors threw 
the toga round his head and thus dragged him away, whereby 
they often nearly strangled him, the blood flowing from his 
mouth and nose. A person dragged in this manner endea- 
voured of course to defend himself by drawing the toga towards 
himself; the lictor then took a knife and cut a hole in the toga 
through which he put his hand and so forced his prisoner 
along. This is expressed by the phrase vestein scindere. But 
the lictors rarely made use of such violence, because the j^eople 
did not easily tolerate it. Volero Publilius being resolute and 
strong, dashed away the lictor, ran among the plebeians and 
called upon the tribunes for assistance. The latter, however, 


being themselves thoroughly terrified, remained silent, where- 
upon he addressed himself to the plebeians : the people rushed 
in a body upon the pursuing lictors who were easily over- 
powered. The young patricians ran to the spot, and a strug- 
gle ensued, in which the tyrants were driven from the forum 
in a very short time. On the following day, the consuls again 
attempted a levy, but were equally unsuccessfiil, and they 
then abstained from making any further trial in the course of 
that year. The murder of Genucius had only rendered matters 
far worse, and Volero Publilius was elected tribune for the 
year following, a clear proof that the sanction of the curies was 
no longer requisite. 

An ordinary man Avould have summoned the consuls of the 
preceding year before the court of the plebes; but this would 
only have been a miserable piece of revenge. Publilius saw 
that the great exaspeiation of the commonalty must be made 
use of to gain permanent advantages for them; and for this 
reason, contrary to the expectation of all, he took a step which 
properly speaking he Avas not allowed to take, but it was the 
beginning of a new order of things. He called upon the 
plebes to declare that they had a right to discuss the affairs 
of the state on the proposal of a tribune, and to pass valid reso- 
lutions; and further that the tribunes should no longer be 
elected by the centuries, but by the tribes. These rogations, 
which are much clearer in Dionysius and Dion Cassius (in the 
abridgment of Zonaras) than in Livy, do not allude to one 
circumstance, viz., that such resolutions of the tribes required 
the sanction of the senate and curies in order to become law; 
it is impossible that the Publilian law should have gone so far 
as to make the same claims as the Hortensian, as is clear also 
from the cases which occur. The development of the states of 
antiquity shows no svich abrupt transitions any more than 
nature herself; and the demands made by the Hortensian law 
would have been inconsistent and senseless in those times. 

The manner in which business was now done was the fol- 
lowing: — The tribunes made their legislative proposals on a 
market day ; for the people, the populus as well as the plebes, 
could not transact biisiness on all days, the curies and centuries 
only on dies comiiialcs, and the tribes only on the nundines ; it 
was the Hortensian law that first empowered the centuries also 
to assemble on the nimdines. The accurate expressions are 


popnlusjubet.plebs scisit ; it was never said plehs jubet ov popu- 
liscituin. The plebes at first met in the forum, but afterwards 
in tlie area Capitolina, the pupulus in the comitium or in a grove 
outside the pomoerium, called the aesculetum or lucus Petelinus. 
In the concilium plebis the votes were given by means o^tabellae, 
and in the concilium of the curies, viva voce. There is no trace 
of its having been necessary to announce by a previous 
promulgation the subject of discussion in the concilium 
populi. The senate had no power to bring anything directly 
before the plebes; it could only commission the consvds to 
have a conference with the tribunes on any question; 
the curies on the other hand could not transact any busi- 
ness without a senafusconsultum, and in their assemblies nothing 
could be done without a curule magistrate or an interrex, 
who were not even allowed to show their faces at the meetings 
of the plebs.^ Now when the tribunes wanted to bring a bill 
before the commonalty for deliberation, they exhibited it in 
the foriun in albo in trinundinum, that is as a matter to be de- 
termined upon after fifteen days, the first nundines being 
included in the reckoning. A concio advocata might take 
place at any time, for the forum was always crowded, and the 
tribune might ascend the rostra and address the people, or give 
an opportunity of speaking to others, especially those who 
intended to speak against his proposal {edocere plebem). But 
such deliberations were only preliminary, not decisive; just as 
when the British parliament forms itself into a committee, in 
which mere resolutions are passed, or as when the French 
chambers have a preliminary deliberation upon a legislative 
proposal in the bureaux; the deliberation on the day when a 
question was to be put to the vote was quite different. It was 
necessary that every transaction of the popidics as well as of the 
plebs should be completed before sunset, otherwise the day 

' In our manuals of antiquities these distinctions are neglected. However 
valuable the cai'lier works on this subject arc in reference to detail, tliey give us 
no assistance in comprehending the political state of Rome. The works of Si- 
gonius and Beaufort deseiwe to be recommended as containing ample materials 
an-anged by ingenious men ; in regard to later times we cannot be grateful 
enongh to them, for the vast amount of information Mhich they afibrd. Tlic 
commentary of Manutius on Cicero's letters is (juite indispensable for any one 
who wishes to understand that period, and his M'ork De Dicbtis is excellent, 
btit as to the earlier times, he too is in tlic dark even more than others. Tlie 
work of Adam is in many respects invaluable, but the first part contains a great 
deal which is incoiTcct. — N. 


was lost; the plcbs had their auspices only in later times, but 
a flash of lightning or any similar phenomenon separated the 
populus (dies diffissus). When a tribune had promulgated his 
rogation in alho fifteen days previously, the decisive deliberation 
took place. We are too apt to represent to ourselves these 
proceedings as tumultuous; the people assembled early in the 
morning, the deliberation lasted the whole day, and one person 
rose after another speaking either for or against the proposal : 
the opposition endeavoured eximere diem, in order that it 
might be impossible to come to a conclusion before sunset: 
which was observed from the steps of the curia Hostilia", and 
then suprema tempestas was announced. In such cases, the 
tribune was again obliged to wait eight days and again to 
promulgate in trinum nundinum. This form must have been 
customary even in the earliest times in all the deliberations of 
the plebes, for there had been plebiscita^ as long as the plebes 

If, on the other hand, the discussion was closed and the 
votes were to be taken, the tribune called upon the patricians 
and clients to withdraw, and as the rostra stood between the 
comitium and the forum, the populus withdrew to the former. 
Hereupon the forum was divided by ropes into a number of 
squares, into each of which a tribe entered, and each tribe then 
voted for itself under the management of its tribune. Wlaen 
it became known that the tribes had passed the resolution, the 
patricians had the right of rejecting it, just as in England the 
house of Lords and the king may reject a bill sent up by the 
house of Commons ; but if the latter is determined to have the 
bill passed, it would be quite impossible to reject it; such a 
measure would be the signal for a dissolution of the govern- 
ment. The patricians would not allow matters to come to such 
a crisis, and therefore usually endeavoured to prevent the 
plebes from coming to an obnoxious resolution. We might 
ask, what advantage there was in preventing a resolution one 
day, since it might be carried the next? A great deal was 
gained; a respite of three weeks, in which perhaps a war 
might arise, which would put a stop to every thing; nay a 

* The discovery of this place has hecn tho key to all my investigations on 
Roman topograjihy. — N. 

' The orthography pkhisscita is quite wrong; plcbi is the ancient genitive of 
plebes ]nst as Hercules, IJeicuU; Cacles, Cacli; dies, dii. — N. 


matter might be dragged on through a whole year, but then 
the evil only increased and the exasperation of the people rose 
higher and higher. This is the folly which all oligarchs will 
be continually guilty of in some form or other as long as there 
are oligarchies. The patricians were blind enough not to see 
that if they could get up among the plebeians themselves a 
sufficiently strong party to oppose a proposal, the consequences 
would be the same as if a resolution were actually passed and 
afterwards rejected, but without any odium being attached to 
them. In the end the patricians never shewed sufficient cou- 
rage to let matters come to a crisis : they always yielded but in 
a hateful manner, and reserved to themselves their ancient 
rights, no part of which they would give up except on 


The great importance of the Publilian law is that the tribunes 
now obtained the initiative; until then it had been quite in 
the power of the senate and the patricians either to allow a 
legislative proposal to be discussed or to prevent it : the consul 
first made his proposal to the senate, and it was only after the 
latter had given its consent to the proposal, that it was brought 
before the curies or the curies and centuries. But as the tri- 
bunes were now at liberty to lay any proposal before the 
commonalty, they thereby acquired the power of introducing 
a discussion upon any subject which required it. There were 
points which urgently demanded a change, and among them 
many of the highest importance, which without the Publilian 
rogation would never have been discussed in a constitutional 
way. The Publilian laws therefore were beneficial, for had 
they not been passed, the indignation of the plebeians would 
have vented itself in another way, and the state would have 
been torn to pieces in wild exasperation; I cannot, however, 
blame the rulers of that time for not seeing the beneficial 
results of the laws. But the angry manner in which they 
opposed the tribunes was as blamcable as it was injurious; the 
mode of their opposition threw the formal injustice upon the 


opposite side, for I cannot deny that tlie Publllian law was 
contrary to tlie existint^ order of things, and an irregularity. 
The senate might have disregarded such a jjlebiscitum altogether, 
or might have declared that the plebes were not qualified to 
pass it; but when the tribunes called upon the populus to with- 
draw from the forum, the patricians refused to go, and with 
their clients spread all over it, so that the plebeians were pre- 
vented from voting ; they drove away the servants who carried 
ihe voting urns, threw out the tablets containing the votes, 
and the like. After this had been attempted once or twice 
more, the exasperation of the plebeians rose to. the highest 
pitch and a fight ensued, in wdiich the patricians and their 
consul Appius Claudius were driven from the forum. The 
consequence was a general panic among the patricians, because 
they saw that it was impossible to resist the infuriated multi- 
tude. But the plebeians did not stop here: they put them- 
selves in possession of the Capitol but without abusing this 
victory, though the tribunes are generally censured. I do not 
mean to represent the plebeians as champions of virtue or their 
opponents as thorough knaves: such an opinion would be ridi- 
culous, but the conduct of the plebeians contains a great lesson ; 
those who in such times have the power in their hands, often 
abuse it, whereas the oppressed are moderate in their conduct, 
as we see especially in the case of religious parties. I believe 
the Jansenists at Utrecht would not have the excellent repu- 
tation which they fully deserve, if they were not the oppressed 
church : it is often a salutary thing for a man to belong to the 
persecuted party. The plebeians used their victory only to 
carry their resolution. Although Appius even now exerted 
all his influence to induce the senate to refuse its sanction, yet 
the senators were too much impressed with the greatness of the 
danger, and the law Avas sanctioned. Livy refers this law 
merely to the election of the tribvmes, but Dionysius and Dion 
Cassius (in Zonaras) give the correct account, f-ivy did not 
clearly sec the peculiar importance of these laws, but at the 
close of his narrative he mentions some poiiits which presup- 
pose what he has not stated. 

Had the patricians been wise, they ought to have been 
pleased at the issue of the affair; no one at least could regard 
it as a misfortune. The repeal of such a law is impossible, but 
instead of .seeing this, the patricians with tlicir weakened 


powers continued attempting to undo what had been done, and 
were bent upon taking revenge. The plebeians still looked 
upon the consul whom they had not elected as illegal, and 
refused to obey him ; in this predicament was Appius Claudius, 
who led an army against the Volscians, and on his march 
began to punish and torture the soldiers for the most trilling 
transgressions. Dionysius' account of these things is very 
credible, and seems to be founded upon ancient traditions. The 
plebeians opposed the consul with stubborn defiance, and would 
rather be punished than obey him. Immediately before the 
battle, they determined to take to flight, and accordingly ran 
back to tlie camp, although the Volscians were not on that 
account the less bent upon pursuing and cutting them to pieces ; 
the Romans did not even remain in the camp, but continued 
their flight till they reached the Koman territory. There Appius 
did a thino- which might seem to us incredible, were it not ac- 
counted for by the influence of the allies^ the Hernicans and 
Latins who were under his command: he put to death every 
tenth man of the army, and led the survivors to Eome. The 
consequence was, that in the following year he was accused by 
the tribunes before the plebes; we may look upon Livy's mas- 
terly description of this as based upon the account of one who 
had a thorough knowledge of the events, though it is more 
detailed than he found it in the annals. Appius displayed the 
greatest defiance and haughtiness, and was resolved not to be 
softened down by entreaties ; even the tribunes allowed them- 
selves to be overawed by hitn. Both our historians agree in 
stating that a respite was granted to him by the tribunes, in 
order that he might make away with himself, — a fact which 
often occurs in the history of Rome, and more rarely in that 
of Greece. He availed himself of the concession even before 
the dawn of the ensuing day, and escaped further prosecution 
by suicide. 

After this, the internal disputes ceased for a time, while the 
wars with foreign nations became more and more important. 
In the year A.U. 286, the Romans conquered Antium, or, ac- 
cording to a more probable account, Antium opened its gates 
to them. In our historians, the town appears as decidedly 
Volscian, and part of its inhabitants are said to have fled to 
the Volscians at Ecetra. I believe that the following is a cor- 
rect view of the matter. AntiiUTi, like Agylla and the other 


coast towns, was originally Tyrrhenian, but there may have 
been a numerous party in the town which, feeling itself too 
weak to resist Kome and Latium, called in the assistance of 
the Volscians; and Ecetra, the south-eastern capital of this 
people, sent a colony to Antium. This colony was looked 
upon by a part of the citizens with hostile feelings, and when 
these citizens called upon the Romans for assistance, the 
Volscian colonists were expelled, and returned to Ecetra. The 
Volscians then attempted to recover what they had thus lost, 
and this save rise to obstinate wars. After Antium had thrown 
itself into the arms of the Romans and their allies, it received 
a colony of Romans, Latins and Hernicans, a remarkable proof 
of the equal manner in which these three nations shared their 
conquests. Every one must see how Dionysius has distorted 
this event; Livy thinks that the Romans who were willing to 
join the colony, did not amount to a sufficient number. Antium 
now was akin to the three allied states, and the ancient Tyr- 
rhenian Antiatans formed the commonalty of the town, while 
the colonists were the burghers; it is probable that each state 
sent 300 colonists, except the Hernicans, who sent 400, for 
among them, the division into four seems to have prevailed, 
whence the mention of the cohortes guadringenariae. The 
Antiates rnille milites, who are met with in the later Volscian 
wars, seem to be these 1000 colonists. Such numerical calcu- 
lations are anything but arbitrary, however much they may 
be opposed to our notions. But the success of the Romans in 
this war was only transitory ; for as they were not the strongest 
in the field, and as the ancient inhabitants always fared ill 
under a colony, it is conceivable that ten years afterwards 
Antium was lost by the Romans, in the same manner as it had 
been gained. According to our division, the establishment of 
the colony at Antium concludes the first period of the Volscian 
war, which henceforth assumes quite a different character. 

The Aequians, who at that time must have been a great 
people (Cicero, in fact, calls them gens magna), seem until then 
to have taken little part in the war ; but the loss of Antium 
roused not only the Volscians of Ecetra to vigorous exertions, 
but also the Aequians. The subsequent misfortunes of the 
Romans are veiled over in our accounts; but the enemy seem 
to have advanced as far as the frontiers of the Roman territory, 
and all the Latin towns were conquered : the Volscians were 


formerly found in the district of Velitrae, but henceforth they 
appear every year on mount Algidus, and obtained possession 
of the arx of Tuscuhim, which was reconquered from them 
only with great difficulty. Several Latin towns disappears 
entirely. Corioli was destroyed: Lavici became an Aequian 
town; Gabii continued to be deserted within its walls as late 
as the time of Dionysius; Praeneste is no longer mentioned, 
and after a period of 100 years, when it reappears in history, 
it was hostile towards Eome; it is probable that only the 
nearest places, such as Tusculum and Lavinium, remained in 
the hands of the Romans. The frontier of the Roman dominion 
was on the other side of the hills of Tusculum ; Circeii, Velitrae, 
Norba and other towns farther East were lost. It is certain, 
therefore, that more than half of Latium was conquered by the 
Aequians who penetrated into it from the Anio, and by the 
Volscians who advanced from the sea coast. 

Some allusions to these events are to be found in the account 
of Coriolanus, for the Romans endeavoured to console them- 
selves for these losses by ascribing them to one of their own 
countrymen, a feeling which is quite natural. In the time of 
the French revolution, I have often seen emigrants rejoicing 
at the victories of the French, although they knew that their 
lives woiild be sacrificed should they fall into the hands of the 
conquerors. In like manner James the second when in exile 
was delio-hted at the victories of the Encrlish. The Romans 
thus fancied that the Volscians lost all their power as soon as 
Coriolanus was no longer with them. But the whole story of 
Coriolanus is neither more nor less than a poem, in which a 
series of events belonging to various years is referred to one 
man and to one period, which events moreover are placed many 
years too early. However hard the Romans may have been 
pressed, it cannot be conceived that neither consuls nor armies 
should have been sent against the enemy, while the latter in 
their victorious career hastened from town to town. It is only 
in the enumeration of the places which were destroyed, that 
we have had a hint as to those wliich became Volscian after 
the destruction of Latium. 

The Volscians penetrated so far, that it became necessary to 
receive men and cattle within the walls of Rome, just as at 
Athens in the Peloponnesian war; and this crowding together 
of men and beasts produced a plague. It is well known that 


great depression always produces a susceptibility for epidemics. 
Jt was the despair of the Attic peasantry, who in the Pelo- 
ponnesian war saw their farms burnt down and their olive 
plantations destroyed, that developed the germs of the epi- 
demic. Physicians of Cadiz have pointed out to me a probable 
cause for the breaking out of the yellow fever which raged 
there in 1800: previously every thing was in a prosperous 
condition, but the despondency which arose from the influx of 
large numbers of poor unemployed people, increased and 
spread the disease with great rapidity. In most cases, the 
germs of an epidemic, though existing, do not come to an 
outbreak, for .particular circumstances are rec^uiied to develop 
them. Thus we may well believe the Romans, that the con- 
flux of jDcople, the want of water and cleanliness (it was in the 
month of August) greatly contributed to produce the epi- 
demic ; but it is probable that the great pestilence which thirty 
years later broke out in Greece and Carthage, began in Italy 
as early as that time. The rate of mortality was fearful : it 
was a real pestilence, and not a mere fever, which alone as 
persons were obliged to sleep in the open air, might, at that 
season of the year, have carried off thousands of people. 
Both consuls fell victims to the disease, two of the four augurs, 
the curio maximus, the fourth part of the senators, and an 
immense number of citizens of all classes, so that sufficient 
conveyances could not be found even to carry the dead to the 
river; the bodies were thrown into the cloacae, whereby the 
evil was increased. During this plague the Volscians and 
Aequians traversed the whole of Latiuni; the Latins offered 
resistance, but suffered a fearful defeat in the valley of Grotta 
Ferrata. In the following year we hear nothing of victories ; 
the disease may have attacked the enemy also, and thus have 
saved Rome. After a few years the plague re-appeared as usual. 

Much of the detail in our accounts of this war is not de- 
serving of notice at all, a great part consists of later inventions 
for the purpose of giving to that dismal period some pleasing 
features. The scene of the wars is mount Algidus, which is 
not a mountain, but a cold interrupted table land several miles 
in circumference between Tusculum and Velitrae ; it forms the 
watershed from which the streams flow partly towards the Liris 
and the Pontine marshes, and partly towards the AnioK "^riie 
' Horace says: Nigraeferacifrondix in Alffirlo. — N. 


district is barren, and in antiquity, as is the case now, it was 
coA^ered witla ever-green stone-oaks; some years ago it was 
the constant haunt of robbers, in consequence of wliich I could 
not visit it, but I have collected accurate information about it. 
There the Aequians and Volscians always appeared and united 
their armies. The same district is the scene of the poetical 
story of Cinciunatus' victories over the Volscians. These 
victories, at least in the form in which they have come down 
to us, belong to a very beautiful poem, and are connected with 
the internal history of Eome, on which account I shall defer 
speaking of them, until I have related to you the commotions 
which occurred after the Publilian law. 


The Publilian law could* not remain without consequences 
destructive of internal quiet, for it was the beginning of a great 
commotion that could not fail to be attended with violent 
shocks. The great subject of complaint with the plebeians 
was the unlimited power of the consuls: they had taken the 
place of the kings; their time of office was limited, but in 
power they were little inferior to the kings, and the conse- 
quences of their undefined power were manifested in the levies 
of troops. As the tribunes had now the right to make legis- 
lative proposals, C Terentilius Harsa first brought forward a 
bill that five men should be appointed to draAV up a law 
respecting the limits of the consular power. It was very 
difficult to"execute this task, for the supreme power can never 
in reality be perfectly defined, and least of all in republics ; it 
must ever be something uncertain so as to be able to act on 
extraordinary emergencies. This circumstance was recognised 
by the Roman republic in the formula videant consides ne quid 
res puhlica detrimenti capiat ; in the earlier times this was quite 
common, and at such junctures it was hardly possible to de- 
termine between the legal use and the abuse of power. The 
task of these five men was of such a nature that we can well 
imagine men of the greatest^ honesty might say much for or 

VOL. I. N 


against the proposal. Some might demand a definition of the 
consular power so as to prevent abuse, while others might 
insist upon the government not being disarmed, in order that 
it might not become powerless in times of danger ; but there 
ought to have been no venom in these differences of opinion. 
It was perhaps intended from the very first that the measure 
should be of a more extensive character, and it may even have 
been intended to divide the consulship equally between the 
two estates. 

During the first year, the commotions were less violent than 
in the next, for according to Dionysius, whose account is quite 
correct, another tribune took up the lex Tei'entilia with this 
extension, that decemvirs, five of the patres and five of the 
plebeians, should be appointed to undertake a general revision 
of the laws. The legislations of antiquity embraced not only 
the civil and criminal law and the mode of procedure, but also 
the political laws and regidations of a temporary nature. The 
legislation of Solon, for example, was a complete reform of the 
constitution, and at the same time regulated temporary matters, 
such as the payment of debts. The notion of the period which 
has just passed away, that general legislation ought to proceed 
from a large assembly of lawyers was quite foreign to the 
ideas of the ancients, who well knew that legislation must be 
the work of a few, and the province of larger assemblies was 
merely to adopt or reject it, the sanction resting with them. 
This is the natural course of things, and hence the ancients 
for the most part followed the maxim that legislation should 
be quite independent of the magistracy: in all the republics 
of antiquity, one man or a few were appointed to make the 
laws, and the people said either Yes or No. Such also was 
the case among the Romans: ten men were to be appointed 
legihus scribendis, who however were to be invested with con- 
sular power. From the remains of the Roman laws, we see 
that each was of great extent, which accounts for the fact, that 
but few persons read the laws, and that most people were quite 
ignorant of their contents : in this respect, the republican form 
in such aifairs is necessarily a mere shew. Dionysius very 
happily expresses himself in saying that the Romans aimed at 
laoyofjbia, and gained la-Tjjopia.^ From an accidental expres- 

' " Properly (in HerodoUts and Thucydides Iffovonia is that state of freedom 
wliero no man is beyond or :ilinve thelaw; it \nr\^\thQv ?i rvpawU nor ahwaar da; 


sion of Tacitus we know that the ancient laws were, for the 
most part, traced to the kings Romulus, Xuma, Tullus and 
Ancus. This shews that each of the three ancient tribes and 
the plebes had their separate laws, which were ascribed to their 
respective archegetes. These tribes and the plebes, which had 
originally been distinct communities, continued to preserve 
their ancient statutes, even after their union into one state. I 
believe that more than a hundred different statutes existed in 
the papal dominions previovisly to the French revolution, and 
many an Italian village containing not more than one hundred 
houses has its own statute or customary law; the late Abbe 
Morelli had collected three hundred different statutes in Italy. 
The same was the case in the middle ages in many parts of 
Germany, though in some instances one and the same law was 
in force over a large extent of country. It is not even certain 
whether the whole of the plebes had the same law, or whether 
a different one was not established in places like IMedullia and 
Politorium ; this hypothesis, it is true, is opposed to the state- 
ment that Servius Tullius abolished all differences among the 
plebeians by dividing them into tribes; but on the other hand 
it is supported by the existence of places like Cameria and 
others, which were Roman colonies and formed separate com- 
monalties. The ancients had a tradition that the clause in the 
twelve tables ordaining that the Fortes and Senates should have 
equal rights, referred to certain places such as Tibur. 

The heads of the plebes might very well insist upon the 
establishment of equal laws for all, an object which was bene- 
ficial not only to them but to the state in general, for the 
disadvantages of such different statutes must have been great 
and keenly felt: the purpose of the reform, therefore, was the 
abolition of every thing which established painful and oppres- 
sive differences between the two orders ; and the tribunes were 
juslified in demanding it. There still was no connubium 
between patricians and plebeians, and the children of mixed 
marriages followed the baser side {deteriorem iiartem sequi). In 
the middle ages, Lombards, Franks, Romans and others lived 
together for centuries in the cities of Italy, each nation having 
its own peculiar laws ; but the inconveniences arising from this 

larryopia (in Demosthenes) is that state where even- free citizen is of equal 
rank." See Hist, of Borne, ii.p.281, note 640. 



circumstance subsequently caused common statutes for all the 
inhabitants of a town to be drawn up. 

The tribunes, however, went further, and as the legislation 
was also to comprise the political law, the legislators were at 
the same time to make a reform of the constitution. The 
Publilian laws had awakened in the nation a life which was 
not in harmony with the ruling power : a new state of things 
was necessarily springing up, which soon found itself in con- 
flict with that which was established. Wliether the patricians 
foresaw to what extent this law would operate, or whether it 
was, from the beginning, intended to be more comprehensive 
than we know, certain it is that they made the most vehement 
opposition to the law, and had recovirse to acts of violence 
similar to those which they had practised before. Kaeso 
Quinctius, a son of Cincinnatus, made himself particularly 
notoi-ious; he repeated all the intrigues of Appius Claudius, 
and heading the young men of his own age and rank and the 
clients, he by violence prevented the plebeians from voting. 
A law {lex Junia) against such violence was passed either 
then or the year before, which declared every one who dis- 
turbed the tribunes in their functions guilty of high treason 
towards the commonalty.^ A person guilty of this crime was 
obliged to find sureties for a sum of money to be fixed by the 
tribunes (the usual number of sureties was ten, each for 3000 
asses), and if he did not await his verdict, the money was for- 
feited to the commonalty. In virtue of this law, the tribunes 
of the year following brought an accusation against Kaeso 
Quinctius before the commonalty. On the trial he was charged 
with having, in conjunction with a band of young patricians, 
maltreated a plebeian so that he died. To us this seems in- 
credible, but it was not so in antiquity : in like manner the 
pentalides in Mitylene ran about with clubs assaulting the 
plebeians of Mitylene. Nay, even in modern times similar 
things have been done: during the minority of Louis the 
Fourteenth such scenes occurred in the streets of Paris, where 
no one dared to walk without arms, there being constant danger 
of an attack. In the time of Queen Anne, there existed a 
band of young nobles in London called Mohocks, who roamed 

* It seems almost inconceivahle that Dionysius should place the passing of 
this law thhty years earlier (a. r. 2G2); his reason perhaps was that Coriolanus 
was said to have been condemned by it. — N. 


through the streets in disguise and attacked the people ; and in 
the reign of King William, Lord Bolingbroke, as we see from 
Swift's correspondence, belonged to such a band. Things of 
this kind could not now occur in any European city, thanks to 
police regulations, which, however much blamed by some 
people, are of incalculable benefit. The accusation excited so 
much exasperation against Kaeso Quinctius, that he did not 
dare to appear before the plebes, but quitted the city. It is 
related, that his father was reduced to poverty in consequence 
of the tribunes having exacted from hirn the sums for which 
sureties had been given. This is impossible; for the tribunes had 
claims on none but the sureties, and if the latter wanted to 
come upon the father, a sponsio must have preceded: even then, 
a man of so noble a family cannot possibly have been de- 
prived of rights which belonged to the meanest of his order; 
he might surely have required his gentiles and clients to indem- 
nify him. The whole account, like so many others, is an 
invention in which a foundation of truth is embellished and 
exaggerated; this making-up of the story might have been 
done with sufficient skill to deceive us, but fortunately it is 
managed so awkwardly that we cannot be misguided. Cincin- 
natus is one of those characters whose names stand very high 
in tradition, but concerning whom the records of history are 
extremely scanty and almost amount to nothing. He after- 
wards appears as consul without anything particular being 
related of him ; it is only in the Aequian war that any striking 
fact is recorded of him. There is a halo of wealth and a halo of 
poverty ; the latter shines more especially in rhetorical times when 
no one wishes to be poor, and when it appears inconceivable that 
a great man should be poor. We may pass over the old story of 
Cincinnatus ploughing his fields, etc. etc. ; but the great enthu- 
siasm which arose from it is a mere interpolation in history. Peri- 
zonius has observed that the same story is related of the dictator 
Atilius Serranus(/e5M/co Serrane serentem),an(]. is therefore quite 
apocryphal: it was probably manufactured out of the name Ser- 
ranus(from severe), which, is surely more ancient than the dictator 
who bore it. The story of Cincinnatus was preserved in a poem 
on his dictatorship, of which the following is the substance. 

A Roman army under the consul Minucius was surrounded 
by the Aequians on mount Algidus; the senate, it is said, 
sent an embassy to Cincinnatus, to offer him the dictatorsliip 


The ambassadors found him on his small farm of four jugera, 
on the other side of the river, engaged in ploughing. Having 
heard the command of the senate, he complied with it, though 
his heart bled at the recollection of the fate of his son. He 
then chose a gallant but poor patrician, L. Tarquitius, for his 
mmjister equitum, and ordered all men capable of bearing arms 
to enlist, every one being required to bring with him twelve 
palisades and provisions for five days. The army broke up in 
the night, and, on its arrival the following morning, marched 
in a column around the Aequian camp; the consul sallied out 
from within, and the Aequians, who were themselves sur- 
rounded by a ditch and palisades were obliged to surrender. 

The whole story is a dream as much as anything that occurs 
in the Heldenhuch. If the Roman army had been in the centre 
and surrounded by an Aequian army, and the latter again by 
a line of Romans, the Romans would have formed a circle 
of at least five miles in circumference, so that the Aequians 
might liave broken through them without any difficulty. I do 
not mean, however, to assert that the dictatorship of Cincin- 
natus is altogether unhistorical, though it is strange that a 
similar event is afterwards related in the sieo-e of Ardea, in 
which the same Cloelius Gracchus is mentioned as commander 
of the Aequians. Cincinnatus now made use of his power to 
get Volscius, who had boi'ne witness against Kaeso Quinctius, 
sent into exile, probably by the curies, for the centuries do 
not seem to have then possessed judicial power. At that time, 
Kaeso Quinctius was no longer living, according to the express 
statement of Livy: he had probably fallen the year before in 
consequence of transactions which shew those times in their 
true colours. After he had gone into exile, the tribunes 
observed symptoms of a conspiracy among the young patricians, 
and there were reports that Kaeso was within the city. It is 
further related that during the night the city was surprised 
from the side of the Carmental gate,whic]i was open, by a band of 
patrician clients, under the command of the Sabine, Appius 
Hcrdonius, who had come down the river in boats. It is 
manifest that it would have been impossible to collect a suf- 
ficient number of boats to convey an army of 4,000 men, 
without its being known at Rome, more especially as the 
Romans were at peace with the Sabincs; and admitting that it 
was necessary to leave the gate open on account of a consecra- 


tion, it must surely have been guarded by double sentinels; 
the enemy, moreover, could not possibly pass the field of Mars 
and occupy the Capitoline hill without being observed. There 
must, therefore, have been treachery. In the night, the people 
were roused from their sleep by a cry that the Capitol was in 
the hands of the enemy, who massacred every one that did not 
join them, and called u]Don the slaves to make common cause 
with them. This naturally created not only the greatest con- 
sternation, but a general mistrust; the j)lGbeians imagined 
that it was an artifice of the patricians who had stirred up their 
clients to take possession of the Capitol, in order thereby to 
intimidate the plebeians; they further believed that, as during 
a tumultus, the consuls would command them to take the 
military oath unconditionally, lead them to a place beyond the 
limits of the tribunitian power, and then require them to 
renounce their rights. The tribunes accordingly declared that 
they could not allow the commonalty to take up arms before 
the laws were passed. We may indeed believe that the 
government was innocent in this affair, but it seems certain 
that there was evidence of a conspiracy in which Kacso 
Quinctius was an accomplice, and that a promise had been 
given to Appius Herdonius to make him king of Kome if the 
imdertaking should succeed. It is not impossible that this 
may rather have been a consj^iracy of the gentes mmores, for 
Ave can still perceive a great gulf between them and the 
majores. When the real state of the case became known, the 
tribunes gave up their opposition and allowed the commonalty 
to take the oath; whereupon the Capitol was stormed under 
the command of the consul. At this time there seems fortu- 
nately to have been a truce with the Aequians, but yet Rome 
was in a most dangerous condition, since no firm reliance 
could be placed upon the continuance of the truce. The consul 
Valerius, the son of Poplicola, the same who is said to have 
fallen at lake Eegillus, was killed; the Capitol was taken by 
storm, the slaves found there were nailed on crosses, and all 
the freemen were executed. There seems to be no doubt that 
Kaeso Quinctius was among the latter; and this may have led 
his father in the following year to take revenge in a manner 
which is pardonable indeed, but ignoble, by exiling A''olscius 
the accuser of his son. The tribunes of the people are said 
to have prevented this accusation beine; made, a remarkable 


instance of the greatness of their power even at that time ; but 
perhaps, they only afforded protection to the accused, and did 
not allow him to be violently dragged into court. The expres- 
sion patricios coire non passi sunt is not applicable till later times. 
The disputes about the trial lasted for a couple of years, 
Cincinnatus either as consul or dictator (probably the latter), 
refusing to lay down his office until he should have obtained 
the condemnation of Volscius. The latter went into exile; his 
surname Fictor, probably fiom. finger e, is one of the examples 
in which either the name arose from the story, or the story 
from the name: so that the statement, "the plebeian M. 
Volscius Fictor was condemned," gave rise to the story that 
he had given false evidence. 

It is obvious, that Cincinnatus has undeservedly been deified 
by posterity : in the time of the decemvirs and tyrants, he did 
nothing; and twenty years after this occurrence, he acted 
completely in the interest of a faction, and shed the innocent 
blood of Maelius. 


Aftek tlie war of a.u. 296, the history of Home takes a 
different turn. We have no express statements as to the 
circumstances which gave it this new direction; but the con- 
currence of several circumstances leaves no doubt that at that 
time the Eomans concluded a peace and treaty of friendship 
with the Volscians of Ecetra, on condition of restoring Antium 
to the Volscians, so that this town assumed the character which 
it retained for 120 years, that is, till after the Latin war. 
Henceforth then, the Volscians no longer appeared every year 
on mount Algidus, and the Aequians alone continued to be 
enemies, but they were of no importance. From this time the 
Antiatans and Ecetrans took part in the festivals on the Alban 
mount, that is, in the Latin holidays ; this is referred to the 
times of Tarquinius Superbus, but at that time Antium was 
not yet a Volscian town. 

Previously to the year A.u. 290, the census amounted to 
104,000, and nfter the plague, this number was diminished 


only by one eighth, whereas one fourth of the senators had 
been carried off; but the cause of this apparent discrepancy is, 
that the Volscians had been admitted to the right of municipium : 
citizens they were not, and consequently as the census lists 
must have included them, they did not embrace Roman citi- 
zens only. But it is more especially the story of Coriolanus 
that furnishes a proof of this treaty. He is said to have made 
the Romans promise to restore the places which they had 
taken from the Volscians, and to admit the Volscians as isopo- 
lites. Both things were done: Antium was restored, and the 
rights of isopolity were granted. We must either suppose that 
the events recorded of the great Volscian war were transferred 
to this story, or that the episode about Coriolanus formed the 
catastrophe of this war, which was followed by the peace ; that 
is, that Coriolanus really was the commander of the Volscians, 
and mediated the peace between them and the Romans. 

These wars, from a.u. 262 to A.u. 266 belong to the cate- 
gory of impossibilities, and that the history of Coriolanus is 
inserted in a wrong place is perfectly clear. The law against 
the disturbers of the assembly of the people could not have 
been passed previously to the Publilian rogations. If the 
Volscians had appeared at the gates of Rome as early as is 
stated by our historians, no domain land would have been left 
about the distribution of which the consul Sp. Cassius could 
have proposed a law, and there would have been no subject of 
dispute. After the unfortunate Volscian wars, in fact, the 
commotions about the agrarian law really did cease, because 
the matter in dispute no longer existed. Further, if the war 
of Coriolanus in A.u. 262 had been carried on in the manner 
in which it appears in our accounts, the Romans would have 
had no place to restore to the Volscians: whereas after the 
great Volscian war, Rome was in possession of several import- 
ant places and was obliged to restore Antium. Lastly, the 
isopolity which was demanded was actually granted in the year 
A.U. 296, as is proved by the numbers of the census. 

As regards the giving up of Antium, the Roman historians 
say that it revolted; which in the case of a colony is absurd. 
The Roman colony was only withdrawn, and the ancient Tyr- 
rhenian population was left to the Volscians. Nay, the very 
circumstance, in consequence of which the war of Coriolanus 
IS said to have broken out, namely the famine during which 


a Greek king of Sicily is stated to have sent a present of com 
to Eome, points to a later period. After the destructive 
Veientine war in the consulship of Virginius and Servilius, 
the fields around Eome had been set on fire at harvest-time 
and were laid waste also at the following seed-time. In a.U. 
262, Gelo was not reigning at Syracuse, but at the utmost was 
a prince of the insignificant town of Gela. Compared with the 
old Koman annalists, who mentioned the tyi-ant Dionysius as 
the king who had sent the present, Dionysius is very sensible, 
for he proves that that monarch did not reign till eighty years 
later; but Dionysius himself must be severely censured for 
mentioning Gelo. After the Veientine war, indeed, according 
to the more probable chronology, Gelo, or at least his brother 
Hiero, was king of Syracuse, and owing to his hostility to- 
wards the Etruscans, he may actually have had good reasons 
for supporting the Romans. All circumstances therefore point 
to this as the real time. The story of Coriolanus is so generally 
known that I need not give a long account of it. The cause of 
its being transferred to a wrong place was the mention of the 
temple of Fortuna Muliebris, as I have already remarked, but 
this temple certainly belongs to an earHer period : a daughter 
of Valerius Poplicola is said to have been the first priestess. 
Now if it were connected with the history of Coriolanus, his 
wife or mother would undoubtedly have been appointed the 
first priestess, as a reward for their services in behalf of the 

The story runs as follows: — C. or as others name him, Cn. 
Marcius Coriolanus, a very eminent young patrician, probably 
of the lesser gentes (for these are more particularly opposed to 
the plebes), greatly distinguished himself in the wars against 
the Antiatans. He was an officer in the army which the con- 
suls led against the Volscians : the year to which this campaign 
belongs was, of course not mentioned in the poem. The army 
besieged Corioli ; the Volscians advancing from Antium wished 
to relieve the place, but Coriolanus took it by storm while the 
army of the consul Avas fighting against the Antiatans. From 
this feat, he received his surname and acquired great celebrity. 
But while in the war he appears as a young man, he is at the 
same time a member of the senate and at the head of the oli- 
garchic faction. A famine was raging in the city: in contra- 
diction to the plebeian statement that the plebes during the 


secession destroyed nothing, we are told that they had in fact 
laid waste the country; but the whole account is evidently of 
patrician origin, and has a strong party colouring. Various 
but useless attempts were made to procure corn ; money was 
sent to Sicily to purchase it, but the Greek king sent back the 
money and gave the corn as a present : it was perhaps a gift 
from the Carthaginians. The senate, it is said, debated as to 
what was to be done with the corn, and Coriolanus demanded 
that it should neither be distributed nor sold, unless the com- 
monalty would renounce the rights they had lately acquired — 
they were to give up their birth-right for a mess of pottage. 
Another proposal not much more praiseworthy, was that the 
corn should be sold to the commonalty as a corporation, from 
which individual members might afterwards purchase it ; hence 
the plebeians were to pay the purchase-money twice : this plan 
was adopted, but it naturally produced great exasperation, and 
on this occasion it also became known that Coriolanus had 
insisted upon making use of the distress for the purpose of 
abolishing the privileges of the plebeians. Livy relates the 
course of events briefly ; but Dionysius gives a very full ac- 
count of them. According to the former, the tribunes brought 
a charge against Coriolanus as guilty of a violation of the peace ; 
and in this they were fully justified by the sworn treaty of the 
sacred mount. The charge was, of course (thougli Dionysius 
does not see this) brought before the plebes, and Coriolanus 
being summoned before the coiu't of the tribes, had the right 
to quit the country before the sentence was pronounced. A 
person could do this after he had given sureties, but it was not 
done in the way usually supposed. He who had to dread an 
unfavourable issue could not go into exile in the manner de- 
scribed in our manuals of antiquities, but he might wait till all 
the tribes had voted except one, as Polybius says. When the 
majority had decided against him he was condemned; but if he 
had taken up his abode as a citizen of a Latin town, for exam- 
ple, the decision was void, but the sureties, at least in later 
times, had to pay. Livy says that Coriolanus met the accu- 
sation with haughtiness, but that on the day of judgment he 
departed before the sentence was pronomiced. Coriolanus was 
perhaps the first who was allowed to give sureties. The com- 
mon tradition is, that he now went to the Volscians. This is 
true (and up to this point indeed I believe the whole story), 


but his going to Attius TuUus at Antium is apocryphal, and a 
mere copy of the story of Themistocles going to Admetus king 
of the Molossians. He is said io have stirred up tlie Volscians, 
who were quite desponding, to venture again upon the war: 
this is a Roman exaggeration intended to disguise the distress 
which had been caused at Rome by the Volscian arms. It is 
further related that he conquered one place after another; first 
Circeii, then the towns south of the Appian road, and next 
those on the Latin road ; and that at last he advanced even to 
the gates of Rome. This is irreconcilable with what follows: 
Coriolanus now appears at the Roman frontier on the Marrana, 
the canal which conducts the water of the low country of 
Grotta Ferrata into the Tiber, about five miles from Rome. 
The Romans sent to him an embassy, first of ten senators to 
whom he granted a respite of thirty days, and then of three 
more, as the fetiales did when a war was not yet determined 
on ; thereupon priests were sent to him and at last the matrons, 
who moved his heart and induced him to retire. 

All this is very poetical, but is at once seen to be impos- 
sible when closely looked into. Livy makes a curious remark, 
in saying that the fact of the consuls of this year having car- 
ried on a war againstthe Volscians would be altogether unknown, 
if it were not clear from the treaty of Spurius Cassius with the 
Latins, that one of them, Postumus Cominius, was absent, the 
treaty being concluded by Cassius alone. But Livy thinks 
that the glory of Coriolanus, which eclipsed every thing else, 
was the cause of the omission — a valuable testimony ! the an- 
cient traditions then did not state that the consul had anything 
to do with the falling of Corioli, but attributed it to Coriolanus 
alone. Now, as we have before seen, it is not true that Corio- 
lanus received his surname from the taking of Corioli, such 
names derived from conqiiercd places not occurring till the 
time of Scipio Africanus : further, Corioli at that period was 
not a Volscian but a Latin town ; it became Volscian in the 
great Volscian war, which we call the war of Coriolanus, and 
was not destroyed till afterwards. The fact of its being a 
Latin town is clear from the list of the thirty towns which 
took part in the battle of lake Rcgillus, though I admit that 
this list may not have been originally drawn up with reference 
to that battle, but rather to the treaty of Sp. Cassius. The 
name of Coriolanus thus signifies nothing more than the names 


Eegillensis, Vibulanus^ Mugillanus and others, and was derived 
from Corioli, either because that town stood in the relation of 
proxenia or clientela to his family, or for some other reason. 

Nothing therefore is historically known about Coriolanus, 
beyond the fact that he wanted to break the contract with the 
plebeians, and that he was condemned in consequence. His 
subsequent history is equally apocryphal. He was condemned 
as a man who had violated sworn rights (Jeges sacraioe), and 
whoever was guilty of that crime, had accursed himself and 
his family; it is further said, that such persons were sold as 
slaves near the temple of Ceres. How then could his wife and 
childi-en continue to live at Rome, if such a sentence had been 
pronounced upon them ? It is impossible to think of mercy in 
those times. The places against which Coriolanus had made 
war were allied with the Romans, and as whoever made war 
against them was at war with Rome, the Romans ought to have 
marched out against him. Consequently, when he appeared 
before Rome he could no longer offer peace or war, but only a 
truce or terms of a truce; and the Romans, on the other hand, 
could not possibly conclude peace on their own responsibility, 
without consulting the Latins and Hernicans. The old tradi- 
tion goes on to say, that the interdictio aquae et ignis which had 
been pronounced against Coriolanus was withdrawn, but that 
he did not accept the withdrawal, and made demands on 
behalf of the Volscians; but when the matrons had moved him, 
he departed and dropped all the stipulations he had made for 
them. From that moment we find no further trace of him, 
except the statement of Fabius, that up to an advanced age he 
lived among the Volscians, and that one day he said: " It is 
only in his old age that a man feels what it is to live in exile 
away from his country." Others, seeing that the Volscians 
could not have been satisfied with such a mode of actinof, 
stated that they followed him on account of his personal in- 
fluence, but that afterwards, being abandoned by him, they 
stoned him to death on the accusation of Attius Tullus. But 
this was not believed by Livy, because it was contrary to the 
account of Fabius. 

We cannot say that the whole history of Coriolanus is a 
fiction, he is too prominent a person in Roman tradition to be 
altogether fabulous. But as regards the statement that he was 
a commander of the Volscian armies, it must be traced to the 


natural feeling that it is less painful to be conquered by one's 
own countrymen than by foreigners : with such national feel- 
ings, the Eomans pictured to themselves the Volscian war, and 
thus consoled themselves and the Latins for the disgrace of the 
defeat, in consequence of which the Volscians made such ex- 
tensive conquests. In the same spirit, they invented stories 
about the generosity of Coriolanus and about his death. I 
believe that Fabius was right in asserting that Coriolanus lived 
in exile among the Volscians to his old age. The statement 
that Rome was on the brink of destruction is probable, and it 
may be admitted that the description of tlie distress is not quite 
fictitious, but it cannot be denied that the three different em- 
bassies of senators, priests and matrons, are inventions made 
for the purpose of elevating the hero. The two estates mutually 
decry each other in their accounts ; hence the plebeians appear 
from the first quite downcast and the patricians quite proud, 
as if they would hear of no reconciliation with Coriolanus. 

I believe that the truth is very different. At that time there 
still existed a great many who had emigrated with the Tar- 
quins, and they gathered together wherever they found a 
rallying point; now I believe that Coriolanus, after withdraw- 
ing to the Volscians, formed such a rallying point for them. 
As he thus found a small army of Eoman emigrants who were 
joined by Volscians, he marched with them to the Roman 
fi-ontier, not that he imagined he would be able to force his 
way through the gates or walls of Rome, but he encamped 
near it and declared war, just like the persons in Dithmarsch 
who had renounced their country. He first granted a term of 
thirty days, that the senate might consider whether his de- 
mands were to be complied with or not. As the senate did 
not come to a decision^ he waited three days longer — a term 
which a state or general demanding reparation takes to consi- 
der whether he shall declare war, or in what manner he is to 
treat the proposals that may have been made to him. Coriola- 
nus was ixndoubtedly joined by the partizans of Tarquinius, by 
many who had been sent into exile in consequence of crimes, 
and lastly by Volscians. The republic invited him to return ; 
the entreaties of his mother, his wife and the other matrons, 
who implored him, can have no other meaning than that he 
should return alone and not bring with him that terrible band 
of men. He probably answered that he could not return alone 



and forsake his companions. If he had returned, he could 
have done nothing else than set himself up as a tyrant, as waa 
so often the case in Greece with the (fivydSef;, whose return was 
a real scourge to their country^ they being almost under the 
necessity of crushing the party by whom they had been ex- 
pelled. We here see him act in a noble spirit, refusing to 
return in this manner, and rather dismissing his own relatives 
on whom he was obliged to make an impression by renouncing 
his own country: a great man might indeed make such an 
impression in those times. Towards the Volscians he behaved 
with perfect justice, and it is possible that he actually came 
forward as mediator between them and Eome, and prevailed 
upon the latter to give up Antium and grant the isopolity. He 
thus discharged his duty towards those who had received him, 
and Eome gained through him the immense advantage of a 
reconciliation with her most dangerous enemy; the Volscians 
had pressed Eome most severely, and there now remained only 
the Aequians, whom it was easy to resist. The childish vanity 
of the Eomans has so completely disguised this Volscian peace 
that until our own times, no one understood it ; without it, the 
whole history would be incoherent; it saved Eome and gave 
her time to recover her strength ; an opportunity which she 
used with great wisdom. 


It is one of the distinguishing features of the history of 
Eome, that many an event which had every appearance of 
being ruinous was the very means of producing a favourable 
crisis in her affairs. After the plague and the unfortunate war 
of the Volscians, we might have expected to see Eome reduced 
to extremities: the peace with the Volscians was, in the eyes 
of posterity, to some extent a humiliation, and for this reason 
they concealed it; but how wise and advantageous it was for 
Eome under the circumstances, we have already seen ; we may 
assert that through it Eome acquired a power which it Avovdd 
never have obtained even by the most successful issue of the 


war. The destruction of the Latin state virtually did away 
with the equality which was secured by the treaty of 
Sp.Cassius. The common opinion, as found in Dionysius, and 
also in Livy, is, that the Latins were subjects of the Romans, 
and that the war under Manlius and Decius in the year a.u. 
410 (415), was a kind of insurrection. This is contradicted 
by the statement of Cincius in Festus, according to which the 
Latins, in his opinion, ever since the time of Tullus Hostilius 
formed a distinct repul:)lic, and had the supremacy alternately 
with Rome. The truth of the matter is this : from the time of 
Servius Tullius down to Tarquinius Superbus, the Latins stood 
in the relation of equality with Rome, but under Tarquinius 
they were subdued; this state of submission was interrupted 
by the insurrection of Latirmi after the expulsion of the kings, 
but was perhaps restored for a few years after the battle of lake 
Regillus, until at length equality was re-established by the 
treaty of Sp. Cassius. It actually existed for a period of thirty 
years; but when the Latin towns were partly occupied and 
partly destroyed by the Volscians, and when scarcely the 
fourth part of the Latin confederacy continued to exist; this 
remnant could of course no longer lay claim to the same 
equality with Rome as the entire confederacy had done before. 
It can be proved that at the beginning of the fourth century 
of Rome, the Latin towns had ceased to be united by any 
internal bond ; they scarcely had a common court of justice, 
and some towns, such as Ardea, stood completely alone. The 
Latins now again came under the supremacy of Rome, as in 
the reign of Tarquinius Superbus. To distinguish what is 
true for different periods is the only thread that can gviide us 
through the labyrinth of Roman history. Isolated statements 
must be examined with great attention, and not be absolutely 
rejected ; even contradictions are of importance in their place. 
As regards the Hernicans, I cannot say with perfect certainty 
whether they were reduced to the same condition as the 
Latins, though it appears to me very probable. After the 
Gallic conflagration, the Latins again shook off the Roman 
dominion and renewed their claims to equality. This claim 
gave rise to a war which lasted thirty-two, or according to the 
more probable chronology, twenty-eight years, and ended in 
a peace by which the ancient treaty of Sp. Cassius was restored. 
Owing to the consequences of the Volscian war, Rome in the 


meantime enjoyed the advantage of standing alone and bein^ 

In the city of Rome itself the ferment was still great, and 
according to Dion Cassius the assassination of distinguished 
plebeians was not an uncommon occurrence. Amid these 
commotions, the agrarian law and the bill for a revision of the 
legislature were constantly brought forward. It is impossible 
to say who induced the plebes to increase the number of their 
tribunes to ten, two for each class: their authority certainly 
could not be enlarged by this numerical augmentation. At 
the time of this increase, we meet with a strange occurrence, 
which however is very obscure. Valerius Maximus says that 
a tribune, P. Mucins, ordered his nine coUeag-ues to be burnt 
alive as guilty of high treason^ because, under the guidance of 
Sp. Cassius, they had opposed the completion of the election of 
magistrates. The times are here evidently in perfect confusion ; 
for ten tribunes were first elected in the year A.u. 297, and the 
consulship of Sp. Cassius occurs twenty-eight years earlier. 
There are two ways in which wc may account for this tradition : 
these tribunes had either acted as traitors towards the plebes, 
which is scarcely conceivable, as they were elected by the 
tribes: or P. Mucins was not a tribune of the people, or at 
least the sentence was not pronounced by him, but by the 
curies, who thus punished the tribunes for violating the peace. 
There must be some truth in the story since it is mentioned by 
Zonaras also (from Dion Cassius) ; it is not impossible that this 
occurrence is identical with the accusation of nine tribunes 
mentioned by Livy about the time of the Canuleian disputes. 

I shall pass over the insignificant wars with the Aequians 
and Sabines, as well as some legislative enactments, though 
they are of great interest in Roman antiquities, and dwell 
at some length upon the Tcrentilian law; in which the 
tribunes demanded an equality of rights for the two estates. 
It would be highly interesting if we could know the detail 
of the disputes on the Terentilian law; but this is impossible, 
and we have only quite isolated statements to guide us. 
One of them is, that a ti'ireme^ with three ambassadors, was 
sent from Rome for the purpose of making a collection of the 
Greek laws, especially those of Athens. The credibility of 
this account has been the subject of much discussion, and I 
now retract the opinion which I expressed in the first edition 

VOL. I. O 


of my Roman history; 1 had then, like my predecessors, not 
considered that the two questions whether the Roman law was 
derived from the Attic law, and whether Roman ambassadors 
did go to Athens are perfectly distinct. If a person asks : Arc 
the Roman laws derived from those of the Athenians? the 
answer must be decidedly negative. There are only two 
Solonian laws, which are said to be found in the Pandects 
also, but these are quite insignificant and might with equal 
justice be derived from the laws of other nations; it would not 
be difficult to find some German laws which likewise ai^jrcc 
with Roman ones. ]\Ioreover we cannot tell how far the 
national afiinity between Greeks, Romans, and Pelasgians, 
might produce a resemblance in their laws. Nothing that is 
peculiar to the Roman law is found in the Attic law; the 
former is quite peculiar in the law of persons and in the law of 
things. The Greeks never had the law of paternal power as 
we find it at Rome; they never had a law by which a wife in 
marrying entered into the relation of a daughter to her 
husband; and in regard to property they never had any thing- 
like the jus mancipii; the distinction between property acquired 
by purchase and absolute property, between property and 
hereditary possession does not exist in the Attic law; the 
Roman laws of inheritance, of debt and of contracts of loan, 
were perfectly foreign to the Athenians, and the Roman form 
of procedure again was quite different from that of the Athen- 
ians. These points are well known to every one acquainted 
with the Attic orators. The Attic law belono's to a much 


later time, when the ancient forms had already become greatly 
softened and polished; in them we behold a state of civil 
society which was wanting in the very thing which charac- 
terizes the Romans. All that we know of the laws of other 
Greek nations is equally foreign to those of Rome, and if per- 
chance the laws of the states of INIagna Graecia had any re- 
semblance to the Roman, it certainly must have originated in 
their common Italian origin; thus the law o^ ager limitotus in 
the table of Heraclea seems to have been like the law estab- 
lished at Rome. 

From these circumstances, many have concluded that the 
account of the embassy to Greece is not entitled to belief, but 
the case may be looked upon in a different light. There is, 
perhaps, none of us who has not at some time or other, after 


mature consideration, undertaken things whicli liave never 
been accomplished: this may happen to a state as well as to 
individuals. The embassy falls exactly iu the time of Pericles, 
between the Persian and the Peloponnesiau wars, wlien Athens 
was at her highest prosperity, and when the fame of that most 
powerful and wise city had spread far and wide. The fact 
that at a much later period (the age of Cassander) when a 
bust was to be erected to the wisest among the Greeks, the 
Roman senate did not select Socrates, but Pythagoras, was 
quite in the spirit of an Italian nation ; but their setting up a 
statue to Alcibiades as the bravest of the Greeks shews how 
flimiliar Athens was to the minds of the Romans; I may add 
that the\' did not judge incorrectly in regarding Alcibiades 
as the bravest. It may, therefore, after all, not have been 
quite in vain that the Romans sent an embassy to Greece; and 
they appear to have made the proper use of it in regard to 
their political constitution. 

Another tradition respecting this legislation states, that 
Hermodorus, a wise Ephesian, atIio was staying at Rome, was 
consulted by the decemvirs. He is said to have been a friend 
of the great Heraclitus, and to have been exiled from Ephesus 
because he was too wise .^ A statua palliata, which was be- 
lieved to be of him, was shewn at Rome dovm to a late period : 
the tradition is ancient, and Hermodorus was not so celebrated 
as to induce the Romans without any motive to call him their 
instructor. He might act as adviser, as it was the avowed 
object of the legislation to abolish the differences between the 
two estates, to modify the constitution so as to make them as 
much as possible form one united whole, and lastly to effect a 
limitation of the consular imperium. But the civil code was 
not by any means derived from Greek sources; for there are 
provisions in Roman law which it is certain Avere expunged 
from the law of Athens even by Solon ; the criminal code pre- 
sents still greater differences. 

It was from the beginning the intention to appoint a mixed 
commission of legislators. In Livy it seems as if the plebeians 
had entertained the unreasonable idea of chooslno- the leoisla- 
tors, five in number, from their own order exclusively; but 
Dionysius states the number at ten, the intention evidently 
being that five should be patricians and five plebeians. Now 



there is another strange statement in Livy, namely, that the 
plebeians urgently entreated that if a revision of the laws was 
to be undertaken, and the patricians would not allow them to 
take a part in it, the patricians themselves might begin it 
alone, and confer with them concerning the principles only. 
The rational conclusion was therefore come to that the members 
of a mixed commission would be involved in perpetual quar- 
rels, and that it would be better to elect them from only one 
estate, provided the fundamental principles were agreed upon. 
It is, nevertheless, a remarkable fact that all authors concur in 
stating that the obnoxious laws, those which were injurious to 
plebeian liberty, were contained in the last two tables, the 
work of the second decemvirs; the first ten are not blamed, 
they merely granted isonomy, respecting which the parties had 
already agreed, as Appius is made, by Livy, to say — se omnia 
jura summis infimisque aequasse. The laws, which had hitherto 
been different for patricians and plebeians, now became the 
same for both orders, so that personal imprisonment and per- 
sonal security might take place in the case of a patrician also. 

There can be no doubt that the first ten decemvirs were all 
patricians of ancient families, and according to the recently 
discovered consular Fasti their title was decemviri consvlari 
potestate legihus scribendis. They were appointed in the place 
of the consuls, the praefectus urbi, and the quaestors. But are 
Livy and Dionysius right in saying that the tribuneship like- 
wise was suspended? It is incredible: for it would have been 
madness in the plebeians thus to allow their hands to be tied, 
and to renounce the protection of their tribunes; it is not till 
the second decemvirate that we find the plebeians appellationi 
invicem cedentes, and then C. Julius brought a criminal case 
before the people. The tribunes must have said: we agree 
that there shall be ten patrician lawgivers, but the continuance 
of the leges sacratae is to us a guarantee of our rights; — and 
the leges sacratae referred to the tribuneship. The error is very 
conceivable, and undoubtedly arose from the fact that the tri- 
buneship was suspended vmdcr the second decemvirate. If we 
bear in mind that under tlic first decemvirs the tribuneship 
was not suspended, and that the object of their labours was a 
common law for all, every thing becomes clear; all points in 
regard to which there might be a collision of the two estates, 
were reserved for subsequent deliberation. 


But besides this task of establishing a general law, the, com- 
missioners had to settle the constitution on the principle that 
the two estates were to be put on a footing of equality. In 
the projected constitution, two points were agreed upon, namely 
that the tribuneship should be abolished, and that the highest 
power should be given to men of both orders. The last five 
names mentioned by Livy in the second decemvirate are 
plebeian and belong to families which do not occur in the Fasti 
previously to the Licinian law, and afterwards only as plebeian 
consuls; Dionysius expressly recognises three of them as ple- 
beians, and the two others who, it is said, were chosen by 
Appius and the nobles from the lesser gentcs, were likewise 
plebeians, as must be evident to every one acquainted with the 
Eoman gentes; Avhence Livy places them at the end of his list: 
the mistake of Dionysius arose from a confusion of the two 
decemvirates. ^ The first decemvirate represented the decent 
primi of the senate, who were elected after a irpo^ovXev/xa 
of the senate by the centuries ; but the second was a crvvap')(^la 
similar to that of the Attic archons, perhaps occasioned by a 
knowledge of the Attic laws. The second election was quite 
different from the first, the noblest, like the lowest patricians, 
canvassed for the votes of the plebeians (canvassing here 
appears for the first time), so that the election was perfectly 
free. Of these decemvirs six were military tribunes, three 
patricians and three plebeians, and these six were in reality the 
commanders in war; of the remaining four, two must be re- 
garded as invested with censorial power and with that of the 
praefectus z^riz combined with the presidency of the senate; the 
other two who had the authority of quaestors, had likewise in 
certain cases to perform military functions. One in each of 
these two pairs, of course, was a patrician and the other a ple- 
beian. Now when Dionysius read that there were three patri- 
cian and three plebeian military tribunes, he might easily 
overlook the fact that the remaining four were likewise equally 
divided between the two orders, especially as the ancient books 
were probably written in a language which was very unintel- 

* As long as I see such an error, and cannot rationally explain it, except on 
the supposition that it was committed by the author in a thoughtless moment, I 
feel uneasy; I cannot rest until I discover the source of the error; and I beg of 
you to exercise your minds in the same manner. Most of the errors in Livy 
and Dionysius are not the result of ignorance but of false premises.— N. 


ligible to him. The three decemvirs whom Dionysius recog- 
nises as plebeians arc, Q. PoeteUus, C.Duilius and Sp.Oppius. 

This constitution was intended to remain for ever. We can 
distinctly see what was the task the decemvirs had to perform 
and how they endeavoured to do it. The distinction between 
the gentes majores and minores disappears from this time. The 
legislators considered the state from the point of view of the 
government, and they reasoned thus: " Since the Publilian law 
the state has been unfortunate ; the tribunes have the power of 
discussing any subject whether agreeable or not; it is therefore 
a matter of importance to transfer this right of the tribunes to 
the decemvirs, as thereby the plebes too would obtain what 
they could fairly claim, for the plebes and populus must stand 
side by side and yet form one whole. The plebes therefore 
no longer want their tribunes, since they may appeal from the 
patrician decemvirs to the plebeian ones. It is, moreover, fair 
that patricians and plebeians should have an equal share in the 
senate, but the plebeians are to come in gradually until they 
shall have reached a certain number. The two estates must 
be carefully kept apart, yet be endowed with equal powers. 
The former right of the gentes to send their representatives 
into the senate, and the custom of a curia (or perhaps the con- 
suls though their power was much more limited than that of 
the censors of later times) electing a new member in case of a 
gens becoming extinct, are to be supplanted by a new institu- 
tion, and a new magistracy must be created to superintend and 
decide upon the civil condition of the citizens, for example, to 
enrol an aerarius among the plebeians, or to raise the plebeian 
■ nobles to the rank of patricians." These are the principles on 
which the second decemvirs acted in their legislation: the con- 
sequences of these laws, and how little they answered the 
expectations formed of them, we shall see hereafter. 


Scarcely any part of the civil law contained in the twelve 
tables has come down to us; one of the few iiortlons with which 


wc arc acquainted is an enactment of one of the last two tables, 
that there should be no connubium between the patres and 
plebes. This enactment is so characteristic, that we may learn 
from it the spirit which pervaded the whole legislation ; it is 
generally regarded as an innovation, for example, by Diony- 
sius, and by Cicero in his work De lie Publica; but this 
opinion is based upon the erroneous supposition that all these 
laAvs were new, as if previously the Komans had either had no 
laws at all or quite different ones. But it never occurred to 
the mind of the ancients to frame an entirely new legislation; 
all they did was to improve that which had been handed down 
to them by their ancestors. As the intention of the decemviral 
leuislation was to brino; the estates into closer connection and 
to equalise the laws for both, such a separation of the two 
orders assuredly cannot have been an innovation. In the 
middle ages too there is scarcely a single trace of such perfectly 
arbitrary legislation; and as I have been told by Savigny, it is 
not found any where except in the laws of the Emperor 
Frederick II. The opinion of our authors is based on nothing 
but their own conception of a new legislation, and is therefore 
of no authority; on the contrary, it is in the highest degree 
improbable that a separation of this kind, with all its subsequent 
irritation should have been introduced at a moment wdien so 
strong a desire after equalit}^ had been evinced. 

But there are some other points which I do consider as inno- 
vations of the greatest importance, such as the unlimited right 
of making a will, which was established by tlie twelve tables. 
This right was conceded to every pate?' familias, but the later 
jiu'ists introduced most important changes for the purpose of 
limiting this dangerous liberty: it cannot have existed from 
the earliest times. The consequence was a double form of 
making a will, namely in presence of the curies and in pro- 
cinctu, that is before the symbol of the centuries, because they 
represented the exercitus vocatus: before these the testator 
declared his will, and if it was previously to a battle, the 
soldier made this declaration in presence of the army itself; 
Avhen a patrician wanted to dispose of his property, the chief 
pontiff assembled the curies, which had to sanction his will. 
The reason of this lay in the nature of the circumstances. If a 
person left children behind him it was probably not customary 
in ancient times to make a will ; but if he died without issue 


his relatives succeeded to his property; if there were no rela- 
tives the property went to the gentes, and if the whole gens 
was extinct it went to the curia. Formerly when I read in 
the Aulularia of Plautus': Nam noster nostrae qui est magister 
curiae, Dividere argeiiti dixit nwnmos in viros, I used to think 
that it was a pure translation from the Greek, for Euclio repre- 
sents an aerarian, and how does he get into a curia? But the 
-whole relation is purely Eoman : property was left to the curia, 
and this inheritance was divided viritim.^ Here then we have 
a good reason why the sanction of the curies was required. It 
is to be regretted that the leaf in Gains which contained this 
law is illegible. In like manner the plebeians too seem to have 
had their gentilician inheritances, which ultimately fell to the 
tribe, and hence here also the exercitus vocatus, that is the cen- 
turies, had to srive its consent, because a will could not be made 
without the auspices, which the plebeian tribes did not possess. 
Similar regulations concerning the succession to property exist 
to this day in the island of Fehmern, as I learned last summer 
during an excursion. The inhabitants consist of two clans or 
gentes with the laws and manners of Dithmarsch; and if a 
member of these gentes wants to make his will, he must give 
to his cousins {gentiles) a small sum as a compensation for the 
money which in reality belongs to them as his gentiles. In 
Dithmarsch Itself this law has disappeared, nor have I been 
able to discover any trace of it in other parts of Germany, a 
proof that very important and general laws may often disap- 
pear and leave but few and slender vestiges. The curies when 
called upon to sanction a will, were of course at liberty to 
refuse it, but as it was a law of the twelve tables: Paterfamilias 
uti legassit supei' pecunia tutelnve suae rei ita jus esio, it is evi- 
dent that the sanction was only dicis causa. This regulation 
had an incredible influence upon Eoman manners; but it was 
necessary, because the connublum between the two orders was 
not permitted, for even the child of a plebeian by a patrician 
woman could not by law succeed to the father's property; and 
if the father wished to make a bequest to such a child, he 
needed a special law to enable him to do so. "Wlien the con- 
nubium was afterwards established, the freedom of making u 

' i. 2. 29; comp. ii. 2. 2. — Ed. 

'' The nature of the curies liad become essentially altered in the course of time, 
bcc Ilist. of Home, ii. p. 319. — Ed. 

THE LAW or DEBT. 201 

will nevertheless continued to exist, and in the corruption of 
later ages, led to the most disgraceful abuses; the lex Furia 
testamentaria which for good reasons I assign to the period 
about A.U. 450, is a proof that a tendency to abuse was mani- 
fested even at that early time. 

The law of debt likewise must have been contained in one 
of the last two tables, since Cicero describes them as thoroughly 
unfair ; for this was binding upon the plebeians only : the last 
two tables undoubtedly consisted of nothing but exceptions. 

The most important part of the decemviral legislation is the 
jus publicum, a fact which was formerly quite overlooked by 
jurists, who saw in it a code of laws like that of Justinian, 
only very scanty and barbarous. But Cicero and Livy ex- 
pressly call it fons omnis publici privatique juris, and Cicero^ in 
his imitations of the laws of the twelve tables also speaks of the 
administration of the republic. All the institutions, however, 
which continued to exist unaltered, were surely not touched 
upon in the twelve tables, as, for example, the whole constitu- 
tion of the centuries; but we have very few traces of the 
changes in the public law which were introduced by them. 
One of them is the enactment, that no more privilegia should 
be granted, i. e. no laws against individuals, or condemnations 
of individuals. Hence we must infer that previously there 
existed regulations against individuals similar to the ostracism 
at Athens. It is probable, moreover, that the mutual accusa- 
tions of the two orders now ceased, and that the centuries were 
regarded as a general national court. There is indeed no 
express testimony, but, even though it is not possible to an- 
swer for the authenticity of all cases recorded, it is, generally 
speaking, a fact well established by the events of history itself, 
that until then the accusations made by the tribunes were 
brought before the plebes, and those preferred by the quaestors, 
before the curies, but afterwards we hear no more of such accu- 
sations. Accusations before the tribes as well as before the 
curies certainly continue to occur in particular cases, but no 
longer in consequence of an opposition between the two 

The change by which the clients became members of the 
tribes — a fact which afterwards becomes clear — was probably 
made at the same time, for the plebeian tribes, independently 
^ In his work De Legibus. — N. 


of tlieir import as such, were also to form a general national 
division; but though there are some plausible reasons for this 
supposition, it is possible also that the change may not have 
been introduced till 100 or 120 years later. If Camillus was 
condemned by the tribes, wc may perhaps account for it in 
this manner; his tribules certainly are mentioned in the trial. 
Among the wise laws of the twelve tables which Cicero incor- 
porates with his laws, he mentions, with reference to his own 
tumultuous condemnation by the tribes, that a judgment de 
capite civis could be passed only by the comitiatus maximus. We 
certainly cannot assert that previously to the deccmviral legis- 
lation, tlie centuries were not authorized to act as a court of 
justice: I have discovered a formula whicli must belong to an 
earlier period, and probably refers to the centuries as a court 
of justice, and the time will probably come when we shall 
arrive at a positive conviction on this point. If it was so, we 
must suppose that the constitution of the centuries as a court 
of justice took place shortly before the deccmviral legislation, 
for till then the judicia capitis belonged to the curies and tribes. 
The trials of Coriolanus and K. Quinctius did not take place 
before the centuries. If in later times we find an instance of a 
condemnation by the curies, it must be regarded as an illegal 
act of violence. The tribunes accordingly henceforth brought 
a crimen capitis before the centuries, and a mere jnulta before 
the tribes; it often happens too that the person who is con- 
demned goes into exile and loses his franchise. Here we must 
bear in mind the principle mentioned by Cicero in his speech 
for Caeciua, that exile did not imply the loss of the franchise, 
for exile was not a punishment: the loss of the franchise did 
not take place until a person was admitted to the citizenship 
of a foreign state. From this point of view, we must look at 
the condemnation of Camillus, if indeed lie was actually con- 
demned by the tribes and not by the curies, for the latter is 
far more probable. 

In this manner, tlie sphere of the nation as a whole became 
greatly extended, and instead of appeals to the two orders 
separately, there occur scarcely any appeals except those to the 
centuries. The existence of this law sufficiently proves the 
mistake of those who believe that the decemvirs assumed all 
jurisdiction to themselves; the error arose from the belief, that 
as the ancient riglit of appeal to the two estates had been 


abolished, an appeal was now made from one college to another. 
The cases of appeal from the consuls to the people afterwards 
occur very rarely, and even those few instances are extremely 
problematical; the appeal to the assembled court of the com- 
monalty was probably abolished, and according to a natural 
development of the constitution, the tribunes, as the direct 
representatives of the plcbes, stepped into its place, since 
a resolution of the whole commonalty was after all only an 

Other laws likewise which arc mentioned, must perhaps be 
regarded as innovations, as for example that a person who 
had pledged himself for debt, should have the same rights as a 
free man, 

Ever since the battle of lake Eegillus the accounts of Livy 
and Dionysius ai'e, in many years, in perfect harmony with 
each other, important discrepancies occurring but rarely. 
The history of the decemviral legislation also furnishes an 
example of this agreement, but other accounts, small as they 
are in number, do not agree with them at all; hence their 
agreement cannot be quoted as evidence that their state- 
ments contain historical truth, but merely leads us to sup- 
pose that the two historians by chance made use of the 
same sources for this period. The narrative of Livy is 
particularly beautiful and elaborate. The statement that a 
second set of decemvirs was appointed because two tables 
were yet wanting, is foolish; I have already expressed an 
opinion that it was probably intended to institute the decem- 
virate as a permanent magistracy, to abolish the consulship and 
tribuneship, and that the decemvirs of the second year were 
elected not as law-givers^ but as the highest magistrates, and 
Avith power to add two tables to the ten already draAvn up. 
My conjecture, which I here state with tolerable confidence, is 
that these decemvirs Avere not elected for one year only, but 
for several, perhaps for five: we are told that on the ides of 
May they did not lay down their ofl^ce, and this is described 
as a usurpation, ilad this been so, it would have been a true 
Svvaareia in the genuine Greek sense of the word in which it 
is the opposite o? Tvpavvi<i, a distinction unknown in the Latin 
language, although not without example in ancient history.* 

'' The constitutional history of Elis presents ii true counterpart of that of 
Konic. The highest magistrate there was at first appuintcil fur life; even in 


In electing the decemvirs it must have been intended, as was 
the case ever after, that whoever had been invested with this 
office should become a member of the senate; but ten new 
members every year would have caused too great an increase, 
and it seems more easy to suppose that our authors overlooked 
the intention that they should hold their office for more than 
one year, than that the decemvirs arbitrarily prolonged the 
period of their office, a thing which they could not have ven- 
tured to do. 

In the second year, history shows us the decemvirs in the 
possession of all magisterial power; they are said to have kept 
a guard of one hundred and twenty lictors {pa^8o(f)6poi), 
twelve for each as was the custom of all Greek oligarchs ; these 
lictors therefore were to serve a pm-pose different from that of 
the consular lictors : they were to be like the cra>fiaTO(f)vXaKe<i of 
the Greek tyrants. The decemvirs are described by Livy and 
Dionysius as profligate tyrants; but this account must be 
received with the same caution as the stories of most tyrants 
in antiquity, for the greatest monsters in history did not com- 
mit their crimes from a mere love of crime, but generally for 
some purpose. Cicero, moreover, relates that although the de- 
cemvirs did not behave quite as became citizens, yet one of 
them, C. Julius respected the liberties of the people and sum- 
moned a popular court to judge one who was not reus manifes- 
tus. Appius Claudius and Sp. Oppius were the presidents of 
the senate: they administered justice in the city, and were 
probably invested with censorial power. Livy very graphically 
says that the forum and the curia grew silent, that the senate 
was seldom convened and that no comitia were held. This 
was quite natural, for as the tribuneship of the people had 
been abolished, there was no comitia of the tribes nor any one 
to address the people in the forum ; there were no politics to be 
discussed, for the constitution was quite new, and in the civil 
law, too, nothing further was to be done. The senate was 

the Pcloponnesian war the gcntes in Elis were alone sovereign, the surrounding 
territory was in a subject condition, and all power was in the hands of a council 
of ninety men who were elected for life. The people was divided into three 
phylae, and each phylo into thirty gentcs. Afterwards the country population 
obtained the franchise. All Elis was divided into twelve regions, and the nation 
into twelve tribes, four of which were lost in war, so that there remained only 
eight.— N. 


rarely convoked, because the college of the decemvirs could do 
most things by itself; the patricians, therefore, went into the 
country and attended to their estates, many plebeians did the 
same, and there suddenly arose in the city a condition of the 
most profound peace. But the people had been so much accus- 
tomed to excitement that they longed for new commotions; a 
feeling of unhappiness came over them, because every thing 
which had stirred up their minds had now disappeared all at 
once. Whoever like myself, witnessed the period of the 
French revolution knows that great mental excitement becomes 
in the end as habitual and indispensable to man as gambling, 
or any other gratification and excitement of the senses. There 
is no feeling more painful than a sudden and perfect peace 
after a great revolution, and such a transition often becomes 
very dangerous. This was the case in the year 1648, when 
the Dutch had concluded peace with the Spaniards at Munster; 
contemporary writers relate the state of things which followed 
was intolerably tedious, the people became discontented and 
gave themselves up to a dissolute life, disputes arose between 
King William III. and the city of Amsterdam : any question 
however trifling was eagerly taken up in order to have an 
opportunity for giving vent to the passions. A similar state 
of feeling existed in France immediately after the restoration. 
Wherever men's minds are in this condition, ill feeling is 
necessarily produced between the government and the people : 
such was the case at Rome, and the people were dissatisfied 
with their new constitution. Hence even if the decemvirs had 
not been bad, or if Appius Claudius had been the only bad 
one among them, they could not easily have maintained them- 
selves, nor would things have remained quiet. The plebeians 
had been disappointed in those members of their order who 
had become decemvirs; at first the tribunician protection is 
said not to have been missed, but gradually the plebeian de- 
cemvirs began to think it proper to use their power for their 
own advantage, and to share the esprit du corps of the others. 
Thus we can understand how the plebeian Sp.Oppius became 
even more odious than the rest, for he, as well as Appius 
Claudius, reduced creditors to the state of addicti ; such deeds 
had hitherto been done only by patricians. 

Under these circumstances, it must have happened very 
opportunely for the decemvirs that a war with the Aequians 


and Sabines broke out, for they thereby acquired the means of 
occupyin<r the people. It is related that the patriots L. Valerius 
Potitus and M. Horatius Barbatus came forward in the senate 
and demanded that the decemvirs should lay down their power 
before an army was enrolled, but that a majority of the senate 
resolved upon a levy being made at once. But I consider the 
speeches in Livy said to have been delivered on that occasion 
to be nothing but empty declamations, promjDted by the idea 
that the decemvirs had usurped their power. If those speeches 
had been actually delivered, the so called patriots would have 
been traitors to tlieir country, for the enemy had invaded and 
were devastating the Roman territory ; resistance was necessary, 
there was no time for deliberations. Nothing, moreover, would 
have been easier than to levy an army, since tribunes no longer 
existed. The story of L. Siccius, whom the decemvirs are said 
to have caused to be assassinated, has in my opinion little pro- 
bability: it looks a great deal too poetical. All we can do is 
to keep to the fact that two Roman armies took the field, while 
the main army was stationed on Mount Algidus against the 
Aequians. In the meantime a crime was committed in the 
city, of a kind which was of quite common occurrence in the 
Greek oligarchies. Appius Claudius became enamoured of the 
daughter of a centurion, L.Virginius. All accounts agree in 
saying that her death, like that of Lucretia, was the cause of 
the overthrow of the decemvirs; tlie statement is very ancient 
and in no way to be doubted: the rape of women and boys is 
a crime which was very commonly committed by tyrants 
against their subjects; Aristotle and Polybius also expressly 
inform us that the overthrow of oligarchies was often the 
result of such violation of female virtue. Appius Claudius 
suborned a false accuser, one of his own clients, who was 
to declare that the real mother of Virginia had been his 
slave, and that she had sold the inflxnt to the wife of 
Virglnius, who, being herself sterile, wished to deceive her 
husband: this assertion, the accuser wanted to establish by 
false witnesses; and Appius was resolved to adjudge Virginia 
as a slave to his client; but this was contrary to the laws 
of the twelve tables, for if the freedom of a Roman citizen 
was disputed, he could demand to be left in the enjoyment 
of it till the question was decided; only he was obliged to 
give security, as a person's value could be estimated in money. 


This was called vindiciae secundum lihertatem., but Appius 
wanted to assign her contra lihertatem. All the people in the 
forum then crowded around him entreating him to defer judg- 
ment, at least till her father, who was serving in the army, 
could return. Wlien the lictor attempted to use force, the 
number of plebeians in the forum became so great and formid- 
able, that Appius had not courage to abide by his determina- 
tion, but requested the accuser to be satisfied with the security 
until the next court-day ; but in order to crush the possibility 
even of a conspiracy, a court was to be held on the very next 
day. At the same time he sent messengers to the camp with 
orders that the father should be kept in the army; but 
Virginius, whom the betrothed of the maiden and other rela- 
tives had previously sent for, appeared on the next morning in 
tlie forum. The appearance of justice was now lost: if Appius 
allowed the matter to come to a formal investigation, the 
father would have unmasked the lie; for this reason Appius 
declared his conviction that the maiden was the slave of the 
accuser, and ordered her to be led away. The general indio-- 
nation at this procedure gave Virginius courage, and under 
the pretext of taking leave of his daughter and consulting her 
nurse, he took her aside into a porticus and plunged into her 
breast a knife which he had snatched from one of the stalls 
round the forum. The bloody knife in his hands, he quitted 
the city without hindrance, and returned to the camp. The 
soldiers on hearing what had happened, unanimously refused 
obedience to the decemvirs, and both armies united. From 
this point our accounts contradict one another; some state 
that the soldiers took possession of the Sacred Mount, and, as 
in the first secession, of the Aventinc ; but others reverse the 
statement. It is to be observed, that the commonalty then 
had twenty leaders, and consequently was again under the 
protection of its tribunes (phylarchs), who appointed from 
among themselves two men who were to act as presidents and 
negotiate with the rulers who were abandoned by the people 
in the city. The tribuni sacrosancti had been abolished by the 
decemviral constitution, but the tribunes as heads of the tribes 
had remained; and, headed by these, the plebeians were now 
in a more decided state of insurrection against the senate and 
the decemvirs than they had been forty years before; at tliat 
time they had seceded for the purpose of obtaining certain 


riglits, whereas now they were fully armed as for a war. In 
this war the decemvirs would necessarily have been over- 
powered, especially as it is clear that many of the patricians 
also renounced their cause, though, as Livy justly remarks, 
most of them loved the decem viral constitution, because it had 
delivered them from the tribunician power: but still many of 
them, such as Valerius and Horatius, were anxious that the 
ancient constitution should be restored, as they were convinced 
that the tribuneship acted as a salutary check upon the consu- 
lar power. It was accordingly resolved to negotiate with the 
plebes, and peace was concluded. 

We still possess some remnants of different accounts respect- 
ing the fall of the decemvirs : that of Diodorus is quite differ- 
ent from the above ; it might be said to be taken from Fabius 
if it did not contain one strange circumstance. According to 
this account, matters came to a decision much more quickly 
than according to Livy, for peace is said to have been con- 
cluded on the very next day after the occupation of the 
Aventine. According to Cicero, the disruption lasted for a 
long time, nor does he know anything of Livy's statement 
that Valerius and Horatius were the mediators; he mentions 
Valerius afterwards as consul and continually engaged in 
reconciling the parties. These are traces of discrepant tradi- 
tions, although the character of this period is in general quite 
different from that of the preceding one, and truly historical. 
According to a statement of Cicero, the plebeians marched 
from the Sacred ]\Iount to the Aventine, which is certainly 
wrong, for they were always in possession of the Aventine ; it 
is moreover probable that the obscure Icilian law referred to the 
fact, that the Aventine should be excluded from the union with 
Rome, and, as the real seat of the plebeians, should have its own 
magistrates. We must therefore suppose the meaning of the 
account to be that the army first occupied the Sacred Mount, 
and then marched toAvards the city where they united with the 
members of their own order on the Aventine. The Capitol 
was surrendered to the armed troops, and this surrender shows 
most clearly the difference between the present plebeians and 
those who had seceded forty years before; the plebeians had 
gained a complete victory. 

The decemvirs laid down their office, and the first election 
was that of ten tribunes which was forthwith held xmder the 


presidency of the pontifex maximus, which was the strongest 
recognition on the part of the patricians; the inviolability of 
the plebeian magistrates thus became scciired by the ecclesias- 
tical law. It is a highly remarkable anomaly that they held 
their concilia in the place afterwards called the Circus Flaminius, 
which was to the plebeians what the Circus Maximus was to 
the patricians. These things happened in December, and 
henceforth the tribunes regularly entered upon their office in 
that month. For the purpose of restoring order in the state, 
it was resolved that two patrician magistrates should again 
be elected, but no longer with the former title of praetors but 
with that of consuls, as we are informed by Zonaras. 


The very fact of the title of praetor being changed into 
consul is a proof that the magistracy was looked upon as 
something diiFerent from what it had been before : its 
dignity had diminished, for praetors are those who go before 
or have the command, whereas the word consuls signifies 
colleagues merely, and is quite an abstract name like 
decemvirs. This new form of the consulship, however, was 
not by any means intended as a restoration of the old con- 
stitution, or to take the place of the decemvirate, but was 
only an extraordinary and transitory measure. As a proof 
of this I may mention that the law which declared any one 
who did violence to a tribune or aedile an outlaw, was now 
extended also to judices and decemvirs. This law has been the 
subject of much dispute, but the mention of the decemvirs 
in it is well authenticated. Even the great Antonius Au- 
gustus, bishop of Taragona, a man very distinguished for 
his knowledge of ancient monuments and public law, but 
who notwithstanding his great historical talent was unfor- 
tunately wanting in grammatical accuracy, saw that the judices 
here mentioned were the centum viri, or the judges who were 
appointed by the plebeians, three for each tribe, to decide in 
all questions about quiritarian property. He mentions this 

VOL. I. P 


merely in a passing remark; but I have fully proved it in the 
new edition of my history. Most people understood these 
judices to be the consuls, and therefore concluded that the 
consuls were inviolable; it was just as great a mistake to 
imasrine that the decemvirs mentioned in the law were the 
decemviri stlitihus judicandis, who did not exist till the fifth 
century of Rome : the decemvirs are undoubtedly the decemviri 
consulari poiestaie, and especially the plebeian ones, the patri- 
cians being already sufficiently protected by their ancient laws. 

When the tribimeship was restored, the patricians may have 
.'i^aid: *' You were right, for the praetors, as they formerly 
existed, had too excessive a power, and hence we shared the 
decemvirate with you ; but now as you have your tribunes 
again, you would acquire an overwhelming power, and you 
must therefore leave the decemvirate to us alone." This the 
plebeians refused to do ; and this put an end to the discussions 
about the restoratioii of the decemvirate ; the consular power 
was retained, but with an important change. According to 
very authentic accounts, the elective assembly down to the year 
A. U. 269 was in possession of a truly free right of election; 
but after this time a change was made, first by a usurpation of 
the curies, and afterwards by a formal contract that one of the 
consuls should be nominated by the senate and sanctioned by 
the curies, and that the other should be elected by the centuries. 
In this election, the centuries might act with perfect freedom, 
as in all their other transactions, which was probably the con- 
sequence of the deccm viral legislation ; but the consul elected 
by them still reqiiired the sanction of the curies. 

The power of the tribunes too was changed in one point. 
Before this time all things had been decided in their college by 
the majority, but according to Diodorus it now became law 
that the opposition of a single tribune could paralyze the whole 
college: this opposition was equivalent to an appeal to the 
tribes, and was an exemplification of the principle vetanfis 
major potcstos. According to Livy this law had existed before ; 
but it is probable that it was at least not recognized until now, 
when the relation of the tribunes to the commonalty was changed: 
they were no longer the deputies, but the representatives 
of their order, which was in reality a change for the worse, 
though its evil consequences were not felt till several gene- 
rations later. Here we again perceive the skill and prudence of 


the government, since they might hope always to find one at 
least in the college, ready to support their interests. Cicero 
says, that the tribuneship saved Rome from a revolution ; if 
the people had been refused their tribunes, it would have been 
necessary to retain the kings. 

The centuries had now obtained jurisdiction; according to 
the religious law, the comitia of the centuries had their 
auspices, the gods being consulted as to whether that which 
was to be brought before the comitia was pleasing to them. 
Now as the tribunes had the riojht to brinsf acciTsations before 
the centuries, it follows that they must have been entitled to 
take the auspices {de coelo ohservare). This is expressed in the 
statement of Zonaras, that the tribunes received permission to 
consult the auspices. According to a remark in Diodorus, any 
person should be outlawed who caused the plebeians to be 
without their tribunes. At the close of the year we meet with 
the strange circumstance of two patricians being among the 
tribunes ; they were either patricians who had gone over to the 
plebeians, or the patricians acted upon the principle, which 
was perfectly correct, that the tribunes, considering their 
power of interfering in the movements of the state, were no 
longer the magistrates of a part of the nation, but of the 
whole nation. It is expressly attested, that at that time many 
patricians went over to the plebeians, but the other explanation 
also has great probability. From this time forward patricians 
are often mentioned as tribules of the plebeians, and in the 
discussions about the separation of the plebcs and their settHng 
at Veil, we read that the senators went about prensantes suos 
quisque tribules ; and about fifteen years after the time of the 
decemvirs, Mamercus Aemilius is said to have been struck out 
from the list of his tribe, and to have been placed among the 
ae7'arii; Camillus too is stated to have applied to his tribules, 
though here, it might be said, we must understand his patri- 
cian gentiles. That in the time of Cicero all the patricians 
belonged to the tribes, is well known, Caesar belonged to the 
tribus Fabia, and Sulpicius to the tribus Lcmonia. After the 
Hannibalian war, C. Claudius is made to say by Livy, that to 
strike a person from all the thirty-five tribes was the same as 
to deprive him of the franchise; and M. Livius removed his 
colleague Claudius from his tribe. More examples of the same 
kind might easily be accumulated. In the early times, there 

p 2 


were both patrician and plebeian tribes, but at a later period 
the three patrician tribes of the Ramnes, Titles and Luceres 
are no longer spoken of, and they appeared in the centuries 
only as the sex suffrogia. The whole Roman nation therefore 
was now comprised in the tribes. The same was the case at 
Athens, when the ten phylae of the demos became the only 
ones, and the four ancient mixed phylae disappeared. I 
formerly believed that this was the work of the decemviral 
lefxislation : but if we consider the care with which the decem- 
virs kept the two orders apart in other respects, we cannot 
possibly suppose, that they introduced a fusion in this particular. 
We must place the change somewhat later, and the fittest 
opportunity seems to be the time of the second censors, so that 
the change was made soon after the decemviral legislation. 
We read in the fragments of Dion Cassius, that the patricians 
preferred the condition of the plebeians to their own, because 
they had greater power, and that for this reason they went over 
to them. The power indeed of the plebeians at that time was 
not greater, but they had greater strength, and it could easily 
be foreseen to what, in the course of time, they would attain ; 
iriany therefore may have thought it a more agreeable position 
to stand in the ranks of those who were advancing, than among 
those who were stationary, and could act only on the defensive. 
The decemvirs were accused; Appius Claudius and Sp. 
Oppius died in prison. The latter was a plebeian, a proof 
that the plebeians must not be regarded as persons possessed 
of peculiar virtues. Wherever a state is divided into factions 
and the ruling party abuses its power, our sympathies go with 
the weaker party. Sp. Oppius was perhaps one of those who 
had before been very loud in his denunciations against tyranny, 
but afterwards became a tyrant himself. L. Virginius, Avho 
had been appointed tribune to avenge the blood of his daughter, 
brought a capital charge against Appius*, and by virtue of his 
tribunician power ordered him to be thrown into prison. 
Livy's account here leads us to a curious point. It is a very 
general opinion that every Roman citizen had the right to 
escape from a sentence of death by going into exile. If this 
had been the case, we might wonder why the punishment of 

' A.Vir<,'inins in Livy is iirol)iibh' n niist.ikcof a copyist wlio was tliinkingof 
the earlier trilninc of fliis Tiamc. — N. 


death was instituted at all, and yet the ancient Roman laws 
were not sparing of it ; but the fact is different : the views of 
the ancients in regard to criminal law differ from ours almost 
more than in regard to any other subject. According to our 
notions a criminal must be tried, even if he has been caught 
in the act; we consider it almost a duty on his part to deny 
his crime, and he must be convicted by evidence ; advocates 
may defend him and attempt to misguide the covirt. Of such 
a mode of proceeding the ancients had no idea: when a person 
had committed a crime, the statement of witnesses was suffici- 
ent to cause him to be forthwith apprehended and dragged 
before a magistrate: if the crime was not a delictum manifestum, 
the offender, if a plebeian, might call for the assistance of a 
tribune and give security; if after this he was set free, he 
might sacrifice his sureties and go into exile. But if he had 
been caught in a delictum manifestum in flagranti, and the testes 
locupletes declared that they were present and bore witness to 
his identity, no trial took place: the criminal was dragged 
obtorto collo, the toga being drawn over his head, before the 
magistrate, who forthwith pronounced sentence. If the day 
on which the criminal was caught, was not a court-day, he was 
taken to prison until the next court-day. If, on the other 
hand, a person committed a capital offence of such a kind that 
catching him in flagranti was impossible, nevertheless the 
accuser had the means of obtaining the imprisonment of the 
culprit.^ Appius Claudius, for example, was gviilty of a capital 
offence: he had deprived a citizen of his liberty, and Virginias 
accused him without allowing him to give security, in order 
that he might not escape ; in such a case the accuser might 
offer to the defendant a sponsio a kind of wager, consisting of 
a sum of money (sacramenjum) on the part of the accuser 
against the personal liberty of the defendant. The accuser 
said : You have deprived a citizen of his liberty ; the accused 
denied the charge, and if the judex chosen for the case declared 
for the accuser, no further trial was necessary; the criminal 
was forthwith led before the magistrate, and executed; if 
however the judex decided against the accuser, the latter lost 
the sacramentum. If the accused declined to accept the sponsio, 
he was thrown into prison. The question now is, whether in 
such a case as this the accuser was obliged to drop his suit or 
"^ On this subject comp. Hist, of Rome. vol. ii. p. 370, ibl.— Ed. 


to accept the security. The passages which decide this question, 
occur in Livy and Cicero. The accused remained in prison 
only till the next court-day, and thus the smallness of the 
prison at Komo becomes intelligible ; confinement in its dark- 
ness was of itself a forerunner of death, and he who was thrown 
into it was lost. Cicero says: carcerem vindicem nefariorum 
ac manifest or um scelerum majores esse voluerunt ; the criminal 
either had his neck broken in the prison, or was led out to be 
executed. The Greek customs connected with imprisonment 
are much more like those of our own times. I may here add 
the remark, that wdien an accusation was brought against a 
filius familias, the fatlier acted as judge; if against a client, the 

Another part of the Koman criminal law entirely different 
from our own, was that relating to offences against the state. 
For many of them no punishment was fixed, it being a dis- 
tinct maxim with the ancients, that the state must preserve 
itself — salus publica svprema lex esto. They well knew that 
the individual crimes against the state admitted of the greatest 
variety of shades, that the same external act might arise either 
from error or from the most criminal intention, and that ac- 
cordingly it was impossible to fix a special punishment for 
each particular case. Hence both Greeks and Romans in all 
judicia publica granted to the accused himself the extraordinary 
privilege of proposing any definite punishment such as he 
thought proportioned to the nature of his offence, and that 
even in cases for which there already existed a precedent. 
The same privilege seems to have been transferred even to 
judicia privata, in those cases for which no provision was 
made in the criminal code. In modern times the foolish 
notion has been established, that a punishment should be in- 
flicted only according to a positive laW; and this sad mistake 
is adojDtcd every where. The ancients followed the directly 
opposite principle: a boy who tortured an animal, was sen- 
tenced to death by the Athenian popular assembly, although 
there was no law for the protection of animals ; it was on the 
same principle that a person who was only guilty of an act 
repulsive to the common feeling of honour, was condemned 
to die. 

Up to this time the patricians seem to have claimed for 
themselves the privilege which exempted them from being 


thrown into prison ; for it is related that Appius called the 
carcei' the domicile of the plebeians. Virginius showed him- 
self generous in granting to Appius time to make away with 
himself But Sp. Oppius was executed, becaiise his crime was 
of a different kind and not one against an individual who 
might be lenient towards him ; for the story that he ordered 
an old soldier who had served for twenty-seven years to be 
scourged, and that this man came forward as his accuser, is 
evidently a fiction. The period of a soldier's actual service 
lasted twenty-eight years, and the introduction in this story of 
one who was in the last year of his military service, is evidently 
a representation of tyranny in general. The other decemvirs 
went into voluntary exile, and their property was confiscated. 
One of them was Q. Fabius, the ancestor of the subsequent 
gens Fabia. After these events, the tribune M. Duilius pro- 
nounced an amnesty for all who had been guilty of any oflence 
during the preceding unhappy period. This precedent is of 
great importance in the history of judicial proceedings among 
Sie Romans. I had distinctly expressed my opinion upon 
these proceedings long before the discovery of Gains, when the 
most absurd notions were current about the Roman criminal 
law ; but the fragments of Gains and the labours of Savigny 
have made everything much clearer. 

At first the patricians had been in great consternation, and 
sanctioned all the laws which were proposed. Among them 
was one which gave to plehiscita the power of laws binding 
upon all, ut quod tributim plebes jussisset popidum tmeret. This 
law is one of the greatest mysteries in Roman history, and 
there is no possibility of giving an absolute historical solution 
of the difiiculty, though I have formed a hypothesis respect- 
ing it, of the truth of which I am convinced. The law as 
stated above is recorded by Livy, who afterwards, in his 
eighth book, says of the second Publilian law ut plebiscita 
omnes Quirites tenerent ; and in the same terms Pliny and 
Laelius Felix (in Gellius) quote the law of Hortensius which 
falls 160 years later, and of which Gaius says: ut plebiscita 
populum tenerent. Now on considering these three laws (the 
Pubhlian is mentioned only by Livy), they seem to enact the 
same thing; but is this really the case? was the law twice re- 
newed because it had fallen into oblivion? If we examine the 
character of these laws in reference to the various times to 


Avhich thoy belong, it will be seen that their meaning was dif- 
ferent, and tliat the force of plebiscita was not interpreted 
always in the same manner. The result of my investigations is, 
that Livy, in mentioning the lex Valeria Horatia, was not 
accurate, because he himself did not see clearly, and because 
he was thinking of the well-known Hortensian law. The law 
probably ran thus: quae plebes tributim jusserit, QUARUM 

RERUM PATRES AUCTORES FACT! 8INT, ut populum teneanf, 

for from this time forward the legislative proceedings are 
often described as follows: when the tribunes had got the 
commonalty to pass a resolution, they then brought it before 
the curies, which forthwith voted upon it; this was an abbre- 
viation of the ordinary mode of proceeding, according to 
wliich legislative proposals, after being sanctioned by the 
senate, were first brought before the centuries and then before 
the curies ; according to the new arrangement, the consultation 
of the senate and the passing through the centuries were abol- 
ished. The change was very important; for now the discus- 
sion of a matter might originate with the plebes themselves. 
It is clear, on the other hand, that without the sanction of the 
curies the plebiscita had not the power of laws, as we see more 
especially during the contest about the Licinian laws; resolu- 
tions of the plebes may at that time have been termed kffes, 
merely because they became leges as soon as they obtained the 
consent of the curies. In cases when the plebes and the curies 
were not divided by party interests, every thing was sanctioned 
by the latter. It must further be observed that this law was 
carried not by a tribunician, but by a consular rogation. The 
Publilian law had been rendered superfluous by the decemviral 
legislation, which did not recognise any comitia tributa. 

The later Publilian law of the dictator Q. Publilius Philo 
has quite a difiercnt meaning; for it dispensed with the assent 
of the curies to a resolution passed by the tribes, because it 
was too tedious a proceeding, and the senate after all had the 
right of proposal. His law ut plebiscita omnem populum teneatit, 
should in all probability run ut plebiscita QUAE SENATU 
AUCTORE FACTA SI NT om7ies QuiHtes TENEANT, for from 
this time it is often mentioned in regard to matters affecting 
the administration, tliat the senate commissioned the consuls 
to negociate with tlic tribunes to bring proposals before the 
tribes; but this occui*s only in matters connected with the ad- 


ministration (y^rrj^la^ara), for example, that a person should 
be invested with an extraordinary imperium, and not in legis- 
lative matters {voixol). This shortening of the proceedings was 
useful : for religious reasons, the curies and centuries could be 
assembled only on certain days, whereas the tribes might and 
did assemble every day, not being restricted by the dies nefasti. 
It became more and more evident, that general assemblies were 
a mere formality and depended too much upon accidental cir- 
cumstances: the supposed personal opinion in voting is onlv 
imaginary; impulse and example do everything. It also be- 
came every day more evident that the more the state increased 
the greater became the want of a regular government; it was 
accordingly of importance to the Romans to devise forms for 
preventing arbitrary proceedings on the part of the govern- 
ment and for preserving publicity. In this respect the Romans 
differed especially from the Greeks, inasmuch as they com- 
mitted themselves with confidence to the personal guidance of 
individuals, which never occurred at Athens. 

Lastly, the Hortensian law has a meaning quite different 
from the preceding laws : it introduced a true democracy, by 
enacting that in the case of legislative measures (for in regard 
to administrative measures the second Publilian law remained 
in force) a preliminary resolution of the senate should be un- 
necessaiy, and that the plebes should have power to pass any 
resolution; the curies were at the same time deprived of their 
functions. This was a decisive victory of the democracy. 
Administrative measures were resolutions on particular emer- 
gencies; and nothing of this kind could be brought before the 
plebes, even down to the end of the sixth century (a.U. 570), 
which had not previously been determined on by the senate; 
but for real laws a resolution of the plebes was sufficient. The 
ancient burghers thereby lost their power of regeneration, the 
balance was destroyed, and the scale sank on the side of demo- 
cracy. The curies had been compelled even by the Publilian 
law, in the year A.u. 417, previously to the meeting of the 
centuries, to declare by a certain formula that they sanctioned 
whatever should be determined upon. It was a misfortune for 
the state that the curies had no means of regeneration : so long 
indeed as resolutions had to pass the centuries, it was not of 
much consequence, but the Hortensian law, which conferred 
all power upon the tribes alone, destroyed the salutary rela- 


tion8 wliicli had hitherto existed, and all the equipoise in the 

In the first stage, these plebiscita were mere resolutions not 
aiFecting the state, but relating to such subjects as, for instance, 
the burial of an important person, the poll-tax, and the like; 
in the second, the plebes by virtue of the first Publilian law 
declared themselves authorized to draw up resolutions on 
general subjects, which however had to be taken into consider- 
ation by the consul, to be laid before the senate, and then to 
pass through the centuries and curies; in the third stage, after 
the Valerian law, a plebiscitum had the force of law as much 
as a resolution of the centuries, and was immediately brought 
before the curies and sanctioned by them. In the fourth, the 
later Publilian law rendered a plebiscitum a sufficient sanction 
of a resolution passed by the senate, which in urgent circum- 
stances, when it was impossible to wait for the next dies corni- 
tialis, was communicated by the consul to the tribunes. It was 
sufficient if the tribunes announced a concilium ; the dies nefasli 
affijcted only curule magistrates and the jmptdus. If for exam- 
ple, at the end of a year an army was in the field, the senate 
would have been obliged to send its resolution to the centuries 
and then to have it sanctioned by the curies; but the shorter 
way now adopted was that the consuls were commissioned, nt 
cum tribunis plebis agerent quam primum fieri posset ad plebem 
ferrent. This does not occur previously to the Publilian law. 
The Hortensian law lastly, in the fifth stage, authorised the 
plebes to act as an independent legislative assembly. 

The consuls now took the field against the Aequians and 
Sabincs, and returned after a brilliant victory, and having 
probably also established a lasting peace with the Sabincs. In 
the meantime the patricians had acquired fresh coura'-'e, and 
those men of their own order, who during the confiision had 
honestly wished to do their best, now became the objects of 
their hatred, and accordingly the senate refused the triumph 
to the returning consuls. This is the first occasion on which 
we sec the overwhelming power of the tribunes, for they inter- 
fered and granted the triumph on their own responsibility; 
their right to do so may be much doubted; but the consids 
accepted the triumph, and if they had been disturbed, the tri- 
bunes would have assisted them. This occurrence shows how 
great the exasperation must have been even at that early period ; 


in the year following it rose to such a height, that, as Livy 
says, the heads of the patricians met and discussed the plan of 
getting rid of their opponents by a general massacre, but this 
senseless scheme, for which they would have had to pay dearly, 
w^as not carried into effect. 

The events which now occurred are very obscure, for the 
piety of posterity has thrown a veil over them. The people 
had got out of the painful stillness which followed the time of 
the decemvirate, but the constitution w^as yet far from having 
found its level, and there were disputes as to who was to govern. 
The plebeians demanded that either the consulship should be 
divided between the two estates, or that the decemviral form 
of government should be restored. In the following year, the 
patricians shewed somewhat more willingness to make conces- 
sions : the quaestores parricidii, hitherto a patrician magistracy, 
were for the first time elected by the centuries ; Valerius and 
Horatius, the consuls of the preceding year, were elected, 
which assuredly was not a mere accident. Many of the ancients, 
as Tacitus, Plutarch, and even Ulpian, aro in error in regard 
to these quaestores, but Gains is right. There were two kinds 
of quaestores, the quaestores parricidii^ who brought accusations 
of offenders against the state before the curies, and the six 
quaestores classici who in books on Roman antiquities are inva- 
riably confounded with the former. Tacitus says of the latter 
what can apply only to the former: " The quaestors," says he, 
" were at first elected by the kings, and afterwards by the con- 
suls, as is clear from a lex curiata of Brutus.^^ But Tacitus 
cannot have seen this lex, for the quaestores parricidii are 
synonymous with the decemviri perduellionis, and the latter 
were always elected by the curies, or rather by the Ramnes and 
Tities which they represented. It is indeed impossible that 
Poplicola caused the quaestores classici, or paymasters, also to 
be elected; but the two who had been formerly elected by the 
curies, and who sixty-three years after the banishment of the 
kings (according to Tacitus), that is, in the second year after 
the overthrow of the decemvirs, were elected by the centuries, 
are the ancient quaestores parricidii, whose office continued 
until it was merfjed in that of the curule aediles. Nine tribunes 
hereupon made the proposal to leave the censorship and quaes- 
torship to the patricians, and either to share the consulship, or 
to institute military tribunes with consular power : only one of 


their colleagues was of a different opinion. It is not impossi- 
ble tliat the stoiy of nine tribunes having once been sentenced 
by the populus to be burnt at the stake, and of one traitor, 
P. Mucius, having carried the sentence into execution, may 
refer to this time.^ In this case the populus means the curiae, 
which again usurped the power of passing such a sentence. 
Among these nine tribunes there was probably a son or grand- 
son of Sp. Cassius, who had renounced his own order, and 
perished in the attempt to avenge his father or grand- 

It was generally wished that the consuls and tribunes should 
be re-elected, but the consuls refused; and Duilius, who had 
been chosen to represent his colleagues, likewise refused to 
accept any votes for the tribuneship. This had evil conse- 
quences, and a division arose: the tribunes who wished to 
remain in office, probably had sufficient influence with their 
friends and followers to cause them to abstain from voting, so 
that only five tribunes were elected, who had to add five to 
their number. It is said that they also chose two patricians, 
which is an argument in favour of our assertion, that not long- 
after the decemviral legislation, the importance of the tribes 
was doubled, inasmuch as they became a general national 

A remarkable change which belongs to this period, is the 
abolition of the law forbidding the connubium between tlie 
patricians and the plebeians. This, as Ave know, had been an 
established custom from the earliest times, and had been incor- 
porated in the laws of the twelve tables. Such a practice is 
usually not repulsive until it is written down among the laws; 
and thus, in this instance too, was raised the storm which 
occasioned the plebiscitum of Canuleius. This is generally 
regarded as the great victory of the plebeians ; for the patri- 
cians, it is said, at last gave way, but reserved to themselves 
other rights. Livy looks upon it as a degradation of the 
ruling order. I will not quarrel with him for saying so, but 
if we look at the matter in its true light, it is evident that the 
existence of such a law injured none more than the patricians 
themselves. Mixed marriages between persons of the two 
estates had undoubtedly been frequent at all times, and as far 
as conscience was concerned they were perfectly legitimate. 

' Hec above, p. 193. 


The son of such a marriage never had the jus ffentilicium, but 
was numbered among the plebeians, the consequence of which 
was, that the patrician order became continually less and 
less numerous. It is an acknowledged fact, that wherever 
the nobles insist upon marrying none but members of their 
own order, they become in course of time quite powerless. 
M. Eehbcrg mentions, that within fifty years one-third of the 
baronial families of the duchy of Bremen became extinct, and 
any body who wished to be regarded as equal to the rest had 
to shew sixteen ancestors. If the plebeians had wished to 
outwit the patricians, they certainly ought to have insisted 
upon the connubium remaining forbidden; and but for the 
Canuleian law, the patricians would have lost their position 
in the state one hundred years earlier. The law was passed, 
but whether it was in favour of the patricians or of the 
plebeians we know not. About such things we cannot speak 
with any probability, for even what appears absurd has some- 
times really happened. 

Afterwards, we once find three military tribunes instead of 
the consuls; and Dionysius on that occasion says, that it was 
determined to satisfy the plebeians by appointing military 
tribunes, three of whom were to be patricians, and three ple- 
beians. But there were only three, and one of them was a 
plebeian. Livy foolishly considers all three to have been 
patricians. He thinks that the plebeians only wanted to have 
the right, but that having gained this they considered them- 
selves unworthy of the oflice, and elected patricians. He 
speaks of the plebeians as if they had been unspeakably 
stupid, thus displaying the confusion of a man, who with all 
his genius, is yet in reality only a rhetorician, and proving 
that he was as little acquainted with the political affairs of 
Rome, as with the regulation of her armies. The probability 
is, that an agreement was made to give up the name of consul 
altogether, since the two orders were no longer separate, and 
to leave the election entirely free between them ; but that, 
nevertheless, all kinds of artifices were resorted to, that the 
elections might turn out in favour of the patricians. In the 
early time, the clients of the patricians were not contained in 
the tribes. They, like their patrons, used to be sent away 
from the forum when the plebeians proceeded to vote, and 
wbocver was not a member of a tribe, was either not contained 


in the centuries at all, or voted in tliem only with the artisans 
and capite censi. But from this time forward there is no 
mention of anything in which the plebes and clients appear as 
opposed to each other, and this ought to convince us how 
authentic our accounts arc, and how little they partake of the 
nature of fables. Is it possible that a late falsifier of history, 
who lived in the seventh century, should have been able so 
accurately to separate legal relations? Such a man is always 
deficient in learning, and even a learned man would have 
blundered here. The clients henceforth appear in the tribes, 
and consequently also in the centuries. This we know, partly 
from express testimony, and partly fi'om the circumstances 
themselves. The discussions of the plebeians now assume 
quite a difierent character; they lose all their vehemence, and 
the contest between two opposed masses ceases .all at once. 
The rejection of plebeians at elections, and the like, no longer 
arose from any external opposition, but from the internal 
dissensions of the plebeians themselves. Formerly the college 
of the tribunes was always unanimous, while henceforth it is 
frequently divided, some of its members being gained over to 
the interests of the senate, and motions wliich used to be 
brought forward by the whole college, are now made by single 
tribunes. These are proofs that the fusion of the two estates 
had been accomplished. 


The military tribuneship had been regarded as a kind of 
compromise. Among the first three, Livy mentions L. Atilius 
Longus and T. Caecilius.^ Instead of the latter, Dionysius, 
in the eleventh book, has Cloelius ; but nothing can be 
decided, since the readings in the eleventh book are all of a 
very recent date. If Caecilius is the correct name, there 
were two plebeians among them ; and this would account for 
the vehemence with which the patricians insisted upon abolish- 
ing the military tribuncsliips. 

' In some modem editions of Livj-, \vc read Cloelins instead of Caecilius, but 
this is an emendation : the MSS. of Dionysius have KKvaiov. — Ed. 


I believe that the censorship was instituted in the same 
year, A.U.311, as the military tribuneships ; and both there- 
fore must have arisen from a common cause, a fact which 
Livy overlooks; and the circumstance of the first censors not 
being found in the Fasti, nor in the libri magistratuum, but 
only in one of the libri lintei, and that as consuls, is accounted 
for by the fact, that the censors were already elected in accord- 
ance with the laws of the twelve tables; and that when the 
patricians carried their point by violent commotion, the cen- 
sors, of whom we have only one trace, were neither consuls nor 
military tribunes, but performed consular functions, and there- 
fore took part in concluding the treaty with the Ardeatans. 
Livy could not explain this, nor could Macer make anything 
of it. It is strange to read in Livy, that the military tribunes 
were obliged to abdicate, because the tabernaculum had been 
vitio captum, and that T. Quinctius, as interrex (more probably 
as dictator), elected the two consuls, L. Papirius Mugillanus 
and L. Sempronius Atratinus, who, however, were not to be 
found in the Fasti; and yet he relates the affair as quite cer- 
tain. It is still more surprising, that the year after he says of 
these first censors, that they were elected censors for the pur- 
pose of indemnifying those quorum de consulatu dubitabaiur, ut 
€0 magistratu parum solidum mogistratum explerent, as if in the 
year A.u. 312 there could have been any doubt as to what had 
happened in 31 L Livy is here guilty of the same thought- 
lessness as when, in the history of the second Funic war, he 
confounds one Heracleitus, a Macedonian ambassador, with 
the celebrated philosopher Heracleitus. 

Now as regards the nature of the military tribunes, it must 
be avowed that this magistracy is very obscure to us.- Livy 
says of them eos juribus et insignibus consularibus usos esse, and 
they are also called tribuni militares consulari potestate ; but 
Dion Cassius, that acute observer, who at one time himself 
occupied the curule chair, states that the military tribunes were 
inferior to the consuls, that none of them ever obtained a 
triumph, although many had done things deserving of one. 
This perfectly agrees with history; we further find that a 
consul was never appointed mogister eguifum, while military 

- The repetitions which occur here and elsewhere arise from the fact, that the 
discussion was interrupted at the close of the hour, and was taken up again at 
the bc'^inninj? of the next Lecture. — Ed. 


tribunes were sometimes invested with that office. This 
seems to show that the military tribunes were not curule 
magistrates, that is according to Gellius' explanation, not such 
magistrates as were allowed to ride in a chariot (as Juno curulis, 
whose image was carried on a chariot); the consuls rode in 
chariots to the curia; the full triumph was called triumphns 
cwulis, according to the monumentum Ancyranum, where the 
number of the triumphi curules of Augustus are mentioned; 
the ovatio was different from such a triumph. ^ It seems, 
moreover, that the military tribunes never had any jurisdic- 
tion ; but it was originally possessed by the censors and after- 
wards by the praefectus urbi, who probably also presided in the 
senate. This latter magistracj'-, too, had been abolished by the 
decemviral legislation, but now appears again. The consular 
power was thus weakened, just as was done afterwards when 
the Licinian law was carried: for when the consulship was 
divided between the patricians and plebeians, the praetorship 
was detached from it and constituted as a separate magistracy. 
It thus becomes intelligible why the plebeians preferred the 
election of military tribunes, even though they were not taken 
from their order, for the power of those magistrates was in- 
ferior to that of the consuls. According to Livy's account, it 
was always the senate which determined whether consuls or 
military tribunes should be elected ; but it is more probable 
that this question was decided by the curies ; confusion here 
may have arisen from the ambiguity of the word patres. The 
military tribuneship, however, presents surprising changes in 
number, for sometimes, though rarely, we find three, more 
frequently four, but from the year A.u. 347 or a.u. 348 regu- 
larly six, wherever they are mentioned, and in one year eight, 
the two censors being included. When there are four, one of 
them usually is the praefectus urbi, so that in reality there are 
only three. The right of the plebeians to be elected military 
tribunes was never disputed, but after the first election it Avas 

' It is a mysterious statement ^vhicll occurs in Livy and elsewhere, that a 
special law was passed for a dictator, ut ei cguum esrcmlcre liceret. This is ex- 
plained by saying that a dictator was not untitled to appeal- on horseback, 
whereas the magister equitum did possess this pri^•ilcg•c. It is possible that the 
dictator was not only entitled to use a chariot, but tliat ho was not allowed to 
ajipear otlierwisc than in a cliariot, especially on his return from battle. An al- 
lusion to this is contained in a verse in Van-o: Dictator ubi currum insedit 
vehitur usque ad oppldum. Oppidum aicordinjj; to Varrois the city wall (also a 
jtlacc surrounded with a wail, in opposition to pagus and vicus). — N. 


nearly always frustrated, though by what means is incompre- 
hensible, for Livy's account, which I have already mentioned, 
is foolish. It is possible indeed that an arrangement was 
made, and that the patricians said: "We grant the institution 
of a weaker magistracy, but then they must be elected from 
among our body exclusively;" or that it was in ancient times 
a privilege of the presiding magistrate not to accept any votes 
{iiomina non accipere) which for various reasons could be re- 
jected; or it may be that when six military tribunes vrere 
elected, the curies conferred the impcrium only upon the patri- 
cians, and refused it to the plebeians. But on this last suppo- 
sition, it is inconceivable how the plebeians should have ac- 
quiesced in it. AVe are here unfortunately without the gui- 
dance of Dionysius, who though he did not comprehend the 
relations, yet gave faithfully what he found in his authorities : 
if we had his account, the whole period would undoubtedly 
be much clearer to us. Bvit we are confined to Livv, and on 
many points we cannot hope to obtain any certain information. 
After the last change, when the number of miHtary tribunes 
became regularly six, we repeatedly find a majority of plebeians 
among them, and the regulation evidently was, that the num- 
ber six should always be complete, and that they should be 
chosen without distinction from both orders. There is every 
appearance, that when this change was introduced, the election 
was transferred from the centuries to the tribes. Everything 
therefore depended upon the honesty of the president, and 
upon his accepting the votes or not. The sad policy by which 
Italy became great in the fifteenth and sixteenth centvu"ies, now 
appears in Roman history, especially in the divisions of the 
college of tribunes; and this is, to some extent, the reason 
why the development of Rome Avas for a time compromised. 

A period in which successful wars are carried on, as was the 
case w^ith Rome from this time down to the Gallic calamity, 
is extremely well calculated to make the subjects of a state 
submit to things which they would not otherwise tolerate. 
The name of the Roman republic w^as surrounded with glory, 
great conquests and much booty were made, the plebeians as 
well as the patricians felt comfortable, and although the rulers 
were not popular, yet things w^ere allowed to go on as they 
were. Rome thus recovered from the decline into which she 
had Slink ever since the rerj'ifncjivm . The grant of the connu- 

A'OL. I. Q 


bium between the two orders also must have exercised a mighty 
influence: the families became more closely connected and 
attached to one another; a patrician born of a plebeian 
mother when sitting in the senate stood on a footing of equality 
with the plebeians, and perhaps did many a thing to please 
them. The number of plebeians in the senate may not have 
been very great, but the mere fact of their being there, even 
without influence, was agreeable to the whole body of 

The censorship being a permanent magistracy, and appa- 
rently the highest, had a lustre which far surpassed that of the 
military tribuneship. If we suppose that it was instituted by 
the twelve tables, it becomes clear why Cicero, in his work De 
Legihus, represents the censors as the first magistracy; he 
probably copied it from the twelve tables, and only omitted a 
few things; for in the twelve tables they had still more attributes. 
In the earlier times the consuls are said to have performed the 
functions of the censors, and this is very probable, considering 
the almost regal power of the" consuls; but it is surprising how 
they can have discharged their enormous duties. The Greek 
states of Sicily and Italy, likewise had their TtfjiTjTai (Athens 
had none), but in no part of Greece were their powers as 
extensive as at Eome. According to the Koman law, the 
censors had to conduct the census, and to determine a person's 
status in society. Accurate lists were kept of the property, 
births and deaths of the citizens, as well as of those who 
were admitted to the franchise. But we must distinguish 
between two kinds of lists. One class consisted of lists of 
persons arranged according to names. Q. Mucins, for example, 
was registered under the tribus Romilia with his name, his 
whole family, and his taxable property. His sons, who had 
the to(/a virilis, probably had a caput of their own. The 
other lists were of a topographical kind, and contained a tabu- 
lar view of landed estates according to the different regions, 
e. g. the tribus Romilia in all its parts. The ancients, in 
general, had much more writing than is commonly imagined ; 
all was done with a minuteness which was part of their poli- 
tical forms. I once saw in London the registers of an Indian 
province — of course in a translation, for I do not understand 
one word of the Indian language — which were drawn up with 
a miniiteness of which we can scarcely form an idea. The 


same was the case with the ancients. The registers of pro- 
perty at Athens were very minute, and so also even in later 
times were the Roman contracts before the curiae. The divi- 
sion of the jugei'a was very accurately recorded in the lists of 
the Roman censors; the caput of each individual contained his 
descent, tribe, rank, property, etc. The censors at the same 
time had the right of transferring persons, for the purpose 
both of honouring and of disgracinof them: but what were 
the offences which the censors punished wdth ignominy 
{ignominia is the real expression)? Every one at Rome was 
expected to answer the definition of his status. A plebeian 
was necessarily a husbandman, either a lauded proprietor or a 
free labourer. This is established by positive testimony, and 
still more in a negative way, for no one could be a plebeian 
who was engaged in craft or a trade. Wlioever so employed 
himself was struck from the list of his tribe, which accordingly 
was not so ranch a personal ignominia, as a declaration that a 
person had passed over from one side to the other. But who- 
ever neglected his farm, was likewise struck out from his 
tribe, i. e- it was declared that he was de facto not a husband- 
man. An eques who kept his horse badly was similarly 
treated, and this was the notatio censoi^ia, by which a person 
was degraded to the rank of an aerarius, being considered 
unworthy to hold his property. An aerarius, on the other 
hand, who distinguished himself, and acquired landed pro- 
perty, was honoured by being registered among the plebeians ; 
and a plebeian who distinguished himself was entered in the 
centuries of the plebeian equites. But the censors certainly 
could not raise strangers to the rank of citizens; for this was 
a point concerning which there were established laws, or else 
the assembly of the people conferred the franchise by an extra- 
ordinary act. A state whose varying elements present great 
differences, where the plebes does not form a close body but 
may complete itself, and contains the aristocratic elements of 
plebeian equites who are not restricted by the census, must 
necessarily have some magistrate for the purpose of assigning 
to every individual his rank; for such an honorary class ot 
men as the equites could not be close or immutable, just 
because it was an honorary class. We may say that the 
power of deciding respecting it might have been left to the 
people; but this would not only have been tedious but also 



perverse, since it might be presumed that the censorSj who 
were chosen from among the most distinguished men, and had 
to bear all the responsibility — one of them having even 
power to oppose the acts of his colleague — would act much 
more fairly and justly, than if the whole people had been 
called upon to decide. The proper filling up of vacancies 
in the senate also required a careful superintendence. It was 
originally an assembly of the gentes, in which each gens was 
represented by its senator: but when gentes became extinct, 
three hundred were taken from the whole body of burghers, 
one hundred from each tribe, so that as gentes became extinct, 
one gens might have several representatives, while another 
might become altogether incapable or unworthy of being 
represented. At a later time the lex Ovinia trihunicia* inter- 
fered, in which it was declared, that out of the whole body 
of the burghers, the wortliicst should be taken without any 
regard to the gentes. If this law belongs to the first period of 
the censorship, it shews that at that time the senate still con- 
sisted of patricians only, and that the worthiest were taken 
from all the three tribes. Tlie account that even Brutus or 
Valerius Poplicola introduced plebeians into the senate under 
the name of conscripti, is a mere fable, or must be regarded 
only as a transitory arrangement. About the time of the 
secession of the commonalty there cannot have been a single 
plebeian in the senate, and their existence there cannot be 
proved till the middle of the fourth century. The senate now 
became a body of men elected by the people, as the magistrates 
obtained the privilege of voting in the senate, and the right of 
being elected into it, when the new list was made up. This 
right extended even to the quaestors. The throwing open of 
the quaestorship to both orders in the year a.U. 346, appears 
to me to have been the first occasion on which the plebeians 
were admitted into the senate ; and Avhen afterwards eight 
quaestors were appointed every 3^ear, the arbitrary power of 
the censors necessarily ceased. They could, indeed, exclude 
plebeians, but the senate consisted of only three hundred mem- 
bers; and as the censors at the close of each lustrum always 
had before them forty men with claims to a seat in the senate, 
it is obvious that the senate might soon become a plebeian 
rather than a patrician assembly. The power of the censors, 
* Festus, s. Y. praetcriti senaiorex. Comp. Hist. Rome, vol. i. note 1 163. — Eu. 



therefore, like that of all the other magistrates, except the 
tribunes, decreased in the course of time. Formerly only a 
censor could sto|) the proceedings of his colleague, but after- 
wards the tribunes also presumed to interfere with the decrees 
of the censors. It was at one time believed to be impossible 
that the censor should have had the powers which were given 
them by the Ovinian law; or if such were the fact, that their 
powers were excessive. Originally, however, they actually 
had great arbitrary power; but as afterwards the two orders 
were no longer exclusively opposed to each other, but the 
government and the people, the latter limited the power of 
the former, and the censors too lost a part of theirs. The 
censorial power did not affect the patricians, for their books 
were closed ; and according to the notions of those times about 
the auspices, no person cotild become a patrician not even by 
adoption, though afterwards cases certainly do occur. 

The question now is : Were the censors allowed to exercise 
their power in regard also to the moral conduct of citizens? 
Could they mark a bad man with a nota censoria ? I formerly 
answered these questions in the negative, excepting perhaps 
cases of decided villany ; but in the recently discovered 
excerpts from Dionysius there is a passage, in which he 
manifestly speaks of the power of the censors to brand every 
moral baseness which could not be reached by the law, such 
as disaffection towards parents, between husband and wife, 
between parents and children, harshness towards slaves and 
neighbours. In the time of Dionysius, it is true, the ancient 
character of the censorship was no longer visible; but this 
is the very reason why we must suppose, that in describing 
the censorship he represented it such as it had been in past 
ages, rather than as it was in his own time, which was known 
to every one. It is therefore probable that the censorial 
power actually had that great extent of which, by our exist- 
ing materials, we can still fix the limits. The censorship 
of Gellius and Lentulus in the time of Cicero was an irregu- 
larity^. Wiiether some tribes were minus honestae, and others 
honesiiores, as early as the period we are now speaking about, 
cannot be determined; but in regard to later times, it is 
acknowledged that the tribus urhanae, and especially the 
Esquilina, were despised^ while the Crustumina stood higher; 
* Cie. />. Clucnt. c. 42 ; Ascon. in Orat. in Tog. Cand. p. 84. Oiclli. — Ed. 


but it would be quite absurd to suppose the same tiling for the 
earlier times. 

The censors were at first elected for a lustrum, or a period of 
five years ; and this seems to have been the period intended 
by the decemviral legislation for all magistrates, according to 
the whole character of that legislation, the principle of which 
was to apply cooling remedies against the political fever, elec- 
tions being always most powerful in stirring up the passions. 
Whether Mam. Aemilius actually limited the censorial power 
to eighteen months, and was therefore branded with ignominy 
by his successors, or whether this is merely a tale which was 
contained in the books of the censors and intended to trace an 
existing law back to some individual, cannot be determined; 
thouo-h it is certain that there existed such books of the censors. 


In the year A.U. 315, a fearful famine broke out at Rome; 
many citizens threw themselves into the Tiber to escape from 
death by starvation. The prices of corn then were in general 
as fluctuating as in the middle ages, which gave rise to much 
-speculation and hoarding up of grain, especially as in Italy corn 
can be kept for a long time under ground. The calamity came 
on unexpectedly; the prae/ectura annonae was then instituted, 
which seems to have been a transitory magistracy: L. Minucius 
Augurinus was the first appointed to the office. He did all he 
could to keep prices down : he ordered the existing stores to be 
opened, compelled the proprietors to sell the corn at a fixed 
price, and made purchases among the neighbouring nations; 
but his measures were too slow, and the means employed for 
the purpose were not sufficient. No effectual help was afibrded 
but by a plebeian eques, Sp. JNIaclius. He at his own expense 
caused large quantities of grain to be purchased in Etruria and 
the country of the Volscians, and distributed the corn among 
the poor. We cannot indeed conceive that his private pro- 
perty could have been very large, but at such times even a 
little aid it^ welcome. A person who conferred such benefits 


upon his fellow-citizens, became easily suspected in the states 
of antiquity of acting from impure motives. Maelius accord- 
ing-ly was accused of trying to gain over the people, and by 
their assistance the tyrannis. Minucius is said to have reported 
to the senate that many plebeians assembled in the house of 
Maelius, and that arms had been carried into it. No man can 
presume to say whether this accusation was well founded or 
not; but at any rate it would have been senseless for a man to 
form a conspiracy, who was not distinguished for anything but 
his wealth, and who would have been opposed no less by the 
tribunes than by the patricians. But however this may be, he 
was regarded as the head of a party, and in order to crush him, 
the senate and the curies appointed L. Quinctius Cincinnatus 
dictator, and he chose Servilius Ahala for his master of the 
horse. In the night Cincinnatus occupied the Capitol and the 
other fortified places, and on the next morning he set up his 
curule throne in the forum, and sent Ahala to summon Maelius 
before his tribunal. ]\Iaelius foresaw his fate, as no tribune 
could protect him against the dictator; he accordingly refused 
to appear, and concealed himself among the crowd of plebeians; 
but Servilius Ahala seized and slew him on the spot. This act 
is much admired by the ancients; but its merit is very doubtful, 
as it may have been a mere murder. The inaefecius annonae, 
according to a very probable account, is stated to have re- 
nounced the patres, and to have gone over to the plebes, and 
to have been appointed the eleventh tribune of the people. In 
a few weeks, it is said, he succeeded in bringing down the 
prices : this shows that the distress had been occasioned by arti- 
ficial means rather than by actual scarcity. The corn contained 
in the granaries of Sp. Maelius was taken by the senate and 
distributed among the people. Moreover, according to Cicero, 
Ahala was charged with murder before the plebes, and went 
into exile : whether he was afterwards recalled, we do not know. 
This also suggests a bad case. The house of Maelius was pulled 
down. The Aequimaelium or place where it stood, was below 
the Capitol, and is now quite buried under rubbish which 
forms a hill at the foot of the Capitol : this point is of great 
importance in Eoman topography.^ 

' The story about Maelius very much resembles one of a Pasha of Aleppo. 
During a great scarcity, he convened all the most distinguislied persons, ordering 
every one to state the amount of corn he pos. esscd. He then rode to their store- 
houses, and on measuring the corn found double the quantity that had been 
returned, and he accordingly look away the surphis, and the dearth ceased.— N. 


When tKe Valerian laws, as we have before seen, so far 
limited the ancient right of the consuls to force the people to 
obedience, that when they pronounced a person deserving of 
corporal punishment^ he could appeal to the commonalty, a 
certain sphere of inflicting punishment not subject to appeal 
was necessarily left to the consuls, for otherwise their authority 
would have been entirely destroyed. This right of punishment 
consisted in the infliction of fines, which regulation also is 
ascribed to Valerius. But this is improbable, for the law of 
the consuls Tarpeius and Aternius, which was passed by the 
centuries, and by which the multa was fixed in heads of 
cattle, as is expressly stated by Cicero {De Re Publico}, is 
framed in terms which are too precise. This could not have 
been the case, if the Valerian law had already determined the 
limitation, unless indeed the rulers had afterwards again been 
guilty of most arbitrary proceedings. I may remark in general 
that all that is said about the Valerii is of a doubtful character, 
as A^alerius Antias looked upon himself as belonging to the 
Valerian gens, and invented a great many things to honour it; 
the Valerii themselves too were vain of popular favour. That 
law fixed two sheep and thirty heifers as the highest multa, 
concerning which Gellius makes a thou2'htless remark, when 
he says that sheep were then so rare that two sheep were equal 
in value to thirty heifers, though immediately afterwards he 
himself mentions the value; that of a sheep as ten, and that of 
an ox as one hundred asses. The fact is simply, that the con- 
suls graduallv increased their fine so as to leave the return to 
obedience open : he who did not appear on the first summons, 
had to give one sheep, if he refused on the second, two sheep, 
tlien a heifer, etc. There is yet another circumstance, which 
we know from Cicero and which shows how little confidence 
can be placed in other accounts: it was not till twenty- five 
years later that the value of these things was fixed in money, 
and at a very moderate rate. Cicero justly regards this as an 
advance in the liberty of individuals. 

The number of quaestors or paymasters who had formerly 
been elected by the king or the curies, and afterwards accord- 
ing to the law of Poplicola by the centuries, was increased 
from two to four, and they were to be partly patricians and 
partly plebeians. At first the patricians prevented the execu- 
tion of this luwj but afterwards the plebeians successfully cstab- 


lislied their claim. This progress "was not merely a matter of 
honour, but a reality, inasmuch as it concerned the immediate 
interests of the plebeians, for they now had a share in the 
administration of the public purse, which accordingly was no 
longer a publicum, but an aerarium. By this means, as I have 
already observed, the senate also was opened to the plebeians, 
and nothing but the censorial power could remove them from 

A flirther progress towards liberty was the fact, that, about 
tAventy years after the decem viral legislation, the right to de- 
termine upon peace and war was transferred from the curies to 
the centuries. That the curies originally possessed this right, 
is established by the testimony of Dionysius, but as the ple- 
beians alone were destined to serve in the ranks and the patri- 
cians deprived them of the booty, it was natural that the 
tribunes should demand for their order the right to determine, 
as to whether they wanted war or not, and the tribvmician op- 
position to declarations of war was nothing but a reservation of 
the rights of the plebes. ^\^ien the centuries had passed a 
resolution to declare war, the curies had of course to give their 
assent; and this they unquestionably always did, as the proposal 
proceeded from the senate, and as it is inconceivable that the 
senate and curies should not have been of one mind. 

The existence of plebeian senators is clear beyond a doubt; 
it is expressly attested that P. Licinius Calvus sat in the senate, 
and hence when an interrex was to be appointed, it was not the 
decem primi alone that met — for through the admission of the 
plebeians they had lost their meaning — but all the patricians 
of the whole senate. This act was termed patricii coeunt ad 
interregem prodendum, and may have been established even by 
the laws of the twelve tables. We can easily understand that 
the Romans might know the laws of the twelve tables by heart, 
and yet not see that there was in them something different 
from what existed afterwards. 

We have now seen how, from tlie time of the decemviral 
legislation down to the taking of the city by the Gauls, inter- 
nal freedom was in a steady process of development, corre- 
sponding to the outward increase of dominion, which shows the 
necessary connection between the two. 

The history of the Italian nations is known to its almost 
exclusively through the Komans; yet if we possessed it, it 


would supply us with the only means of understanding the 
external history of Rome ; for the latter is frequently not only 
defective, but deceitfully corrupted. The decline of the state 
after the expulsion of the kings may have been the consequence 
partly of internal commotions, and partly of the feuds with the 
Latins ; but afterwards the influence of the Etruscans from the 
north gave a fresh blow to Rome, and at the same time the 
extension of the Sabines and their colonies produced a great 
effect. The Romans called the latter Sabelhans, for Sahellus 
is the ordinary adjective along with Sabinus like Hispanus and 
Hispellus, Graecus and Graeculus, Poenus and Poenulus, Romus 
and Roimdus ; it was not till later times that the termination lus 
assumed a diminutive meaning. Sabellus is perfectly equival ent 
to Sabinus, except that, according to common usage, Sabelli 
denotes the whole nation, and Sabini only the inhabitants of 
the small district which bears that name. These extensions of 
the Etruscans and Sabines, then, were the principal cause of 
the decline of Rome, and without them the wars of Porsena 
would not have taken place. If the Etruscans had spread in 
another direction, and if the Sabellians had not been them- 
selves pressed upon and obliged to advance, the Ausonian 
tribes, especially the Aequians, would not have been driven to 
make conquests. 

The period of Etruria's greatness falls in the middle of the 
third century after the building of the city, according to the 
testimony of Cato that the Etruscan colony of Capua or Vul- 
turnum was founded about the year A.M. 260, that is, about 
the time when the Romans were so hard pressed by the 
Veientines. At that time the Etruscans, or, as the Greeks 
call them, the Tyrrhenians, were the most formidable conque- 
rors ; but a crisis took place in the destruction of their navy 
by the Cumssans who were assisted by Hiero, about the end of 
the third century. We can speak of this change only in ge- 
neral terms, for unfortunately all the detail is lost. A mighty 
part of the history of man is here buried in darkness. AlDout 
the same time their power was broken on the Tiber also. 

The Sabines often appear as enemies of the Romans in the 
opposite direction, during the latter half of the third century; 
the earlier accounts of the victories of Valerius over them are 
quite apocryphal. We will not attempt to decide whether 
they were dangerous to the Romans, but there can be no doubt 


that wars were fouglit witli the Sabines as with the other 
tribes in the neighbourhood; all the detail, however, consists 
of poetical fictions. But towards the end of the third century, 
history becomes clearer and clearer, and we can perceive traces 
of the ancient annals. The last Sabine war is that which was 
carried on by Valerius and Horatius in the first year of the 
restoration of the consulship ; it is related too minutely to de- 
serve credit in all its parts, but it is certain, that, during the 
subsequent period of nearly one hundred and fifty years down 
to the time of Curius, the Sabines did not carry on any war 
against the Eomans. This must have had its peculiar reason ; 
and I perceive this reason in a treaty of which not a trace is 
left, but by which isopolity was established between the two 
nations: the existence of that isopolity is attested by Servius 
in his commentary on Virgil. 

About the year a.u. 310, the formation of the Campanian 
people is mentioned, for it is said that at Vulturnum or Capua 
the Etruscans admitted Samnites as epoeci, and shared their 
territory with them. This is a proof of the progress of the 
Sabines in those parts, for the Samnites were a Sabine people. 
The Aequians and Volscians discontinued their attacks upon 
Rome, and the Sabine wars ended : hence we here recognise 
the time when the migration of the Sabines to the south 
ceased, and they left off pressing the Ausonian mountaineers. 
The Etruscans stopped all at once, as is naturally the case with 
a people governed by an oligarchy : when such a people comes 
to a state of rest, it never puts itself in motion again, or ac- 
quires fresh life : at least history furnishes no example of the 
kind. In this manner we may connect the events which the 
Romans have recorded in a very confused manner. 

During the period from A.U. 306 to 323, there was almost a 
total cessation from wars ; the account of the insurrection at 
Ardea, in which the Romans were called upon for assistance, 
has something so strange about it, that we can place no reliance 
on it: we have here a complete repetition of the story of 
Cincinnatus surrounding the hostile army. But in the year 
323 the war broke out afresh and seriously. We do not know 
whether the Antiatans took part in it; but there is no doubt 
that Ecetra did. They then met the Aequians on mount 
Algidus. The Roman armies fought against them between 
Velitrae, which was Volscian, Tusculum, and the Alban 

236 A. rOSTUMlUS tubertus. 

mount: but a battle was lost, wliereupon A. Postumiua 
Tubertus was appointed dictator. This war is perfectly his- 
torical and accurately described; but whether it is true that 
A. Postumius gave weight to his imperium in the minds of 
those who fought under him, by inflexible severity towards 
his own son, may remain undecided; the prevailing opinion 
is, that Manlius followed his example, but from the expression 
imperia Manliana no inference can be drawn; Livy's attempt 
to prove the contrary is, at all events, futile. Postumius led 
the whole strength of the republic and her allies against the 
enemy; he gave one army to the consul, and took the other 
for himself; the former was stationed on the road to Lanuvium, 
the latter on that to Tusculum, near the point where these 
two roads crossed each other. The \^olscians and Aequians 
were stationed in separate camps, one of which was opposed 
by the consul, and the other by the dictator, but the two 
armies were near each other. The enemy attacked the consul's 
camp in the night; but the dictator being prepared, sent a 
detachment to take possession of the Volscian camp, which waa 
almost entirely abandoned, and he himself led the greater part 
of his army to the assistance of the consul, and attacked the 
enemy in the rear. The latter were completely cut to pieces 
with the exception of a small band of men, who, under the 
command of the brave Vcttius Messius, fought their way 
through the Romans. 

This battle is one of those which exert an influence upon 
the history of the world : it broke the power of the Volscians 
of Ecetra and of the Aequians ; the massacre must have been 
immense. The Aequians forthwith sued for peace, and obtained 
it for eight years; from this moment they ceased to be formid- 
able. After this time the Komans were constantly extending 
their dominion, and recovered the places which had been taken 
from them in former wars by the S-^olscians and Aequians. 
Among them we have express mention of Lavici^, formerly 
one of the great Latin towns, Bolae or Bola, Velitrae, Circcii, 
Anxur, Ferentinum, which had formerly belonged to the 
Hernicans and must now have been restored to them, since it 
re-appears among their towns. In this manner, tlie Romans 
advanced as far as the boundaries of Latium in the narrower 
sense, that is just as far as they had penetrated under the kings. 

* Labici, which wc commonly read in Livy, is a mistake of the fomth and 
fit'tii centuries for Laviri. TIic opposite mistake, To/rt instead cf Bola occurs in 
iJic earlv rditi<)n5. — N. 


111 the same way Setia, Norba, Cora, and Signia must have 
been recovered, aud as the Romans and Latins now no longer 
stood on a footing of equality, they must have come under the 
sole dominion of Rome. In the country of the Aequians, the 
Romans advanced to lake Fucinus. The subjugation of the 
Volscians rendered it possible for them to carry on the fearful 
war against Veii. In consequence of these conquests, many 
poor people were provided for, by means of Roman colonies 
established at Lavici and Velitrae; and the colony at Circeii 
was probably made a Latin one. 

After a long interval, the agrarian law was seriously brought 
forward again in the year A. U. 345; it had been previously 
mentioned only once, but slightly. The cause of this silence 
in the preceding years is not quite clear; some assignments of 
colonies took place, but always in conjunction with the Latins 
and Hernicans, and without any beneficial consequences for 
those who did not wish to give up their Roman country and 
franchise. Times of contentment or of discontent in history 
correspond not so much to the political development of rights, 
as to prosperity in general : in happy circumstances man likes 
to enjoy life without thinking much of his political condition. 
Such a period occurred in Germany previously to the thirty 
years' war, when every thing rose in value, and the internal 
condition of the country was tranquil : the same was the case 
in France under Henry IV. Such also, on the whole, was the 
condition of Rome at the time under consideration, and this 
was perhaps the principal reason why no violent internal com- 
motions occurred for so many years. If, however, dviring such 
a period new powers have been developed, then new claims 
also arise, which are put forth at once and with vehemence : 
this was now the case with the agrarian law. Till now, the 
patricians had with great cunning succeeded in excluding the 
plebeians from the honors which belonged to them ; consuls 
were often elected instead of military tribunes, and the number 
of the latter was frequently not complete ; but now the plebeians 
began to insist upon certain claims. The humiliation of Rome 
abroad occasioned by the wars of the Etruscans and Volscians 
had ceased, new conquests quickly raised the city to a great 
height, and under these circumstances the tribunes exerted 
their powers for the advantage of the members of their own 
order. The conquest of Lavici afforded the first opportunity 


for such measures, and its consequences must have been far 
greater than those described by Livj : a colony was demanded 
for that place, but the Roman senate refused it. The Cassian 
law was now never mentioned, but the tribunes brought a new 
lex tnbunicia agraria before the tribes : it was demanded that 
the public land should be divided, and that the portion of it 
possessed by the patricians should again become subject to a 
tax; the latter had originally been a regulation in all agrarian 
laws, but the patricians had contrived to neglect this obligation 
with impunity. These demands of the tribunes were not 
complied with indeed, but they led to the foundation of several 
colonies of Roman citizens, that is purely Roman colonies, 
whence thev are called coloniae Romanae. After the taking of 
Bolae, an unfortunate military tribune, M. Postumius, had 
sold all the booty for the benefit of the publicum {publicum 
redigere^ for the publicum was the separate treasury of the 
curies). This so infuriated the soldiers, that they rose against 
the quaestor and slew him. The military tribune, who tried 
the offenders for this crime, drove them to despair; in conse- 
quence of which they rose against him also, and imbrued their 
hands in his blood, the only instance of the kind that occurred 
before the time of Sulla. The senate treated the matter leniently, 
for the guilt was too evident. The consequences of this insur- 
rection must have been very great, though Livy says nothing 
about them, for from this time forward we never find less than 
six military tribimes, and their election seems to have been 
transferred at that time from the centuries to the tribes, for ' 
otherwise it would be very careless of Livy to speak of a 
tribus praerogativa. The curies, however, still continued to 
confer the imperium upon those who were elected. 

Rome now directed her arms against Veil, which was situ- 
ated at the distance of about twelve miles and measured nearly 
five miles in circumference; its territory must have extended 
to the very foot of the Janiculum. This city was a thorn in 
the side of Rome, which could not become great imtil this 
rival wj s conquered. Fidenae which is called an Etruscan 
city but was really Tyrrhenian, is described from the earliest 
times, and even in the reign of Romulus, as involved in war 
with Rome; it was situated on the Tiber five miles above 
Rome, and had at an early period been occupied by Roman 
colonists, who had been repeatedly expelled but were always 


restored. It was either in A. u. 320 or 329 -^ when the Fide- 
natans again rebelled against the Roman colonists and expelled 
them. We must conceive these colonists as a settled garrison, who 
had their own flirms. Three Roman ambassadors appeared at 
Fidenae to demand reparation and the restoration of the 
colonists. This demand appeared to the Fidenatans so out- 
rageous, that they slew the Roman ambassadors, and threw 
themselves into the arms of Lars Tolumnius, king of Veii ; for 
all the Etruscan towns were governed by kings elected for life. 
Tolumnius marched across the Tiber to their assistance ; and as 
the Romans, after the conquest of the Aequians and Volscians, 
had now become formidable neighbours, the Capenatans and 
Faliscans, two Oscan tribes which had maintained themselves 
in those districts against the Tyrrhenians, likewise came to the 
assistance of the Fidenatans. This army posted itself five miles 
from Rome on the other side of the Anio and created great 
terror in the city. A dictator was appointed, who chose the 
military tribune A. Cornelius Cossus for his master of the horse. 
The Romans fought a successful battle, and Cornelius Cossus 
with his own hands slew the Veientine king, who was charged 
though probably unjustly, with having murdered the Roman 
ambassadors."* After tlris victory, Fidenae was taken and razed 
to the ground, and its territory became ager puhlicus. With 
the Veientines a truce was concluded; and it must have been 
welcome to the Romans to have peace in that quarter, until 
they should have completely broken the power of the Aequians 
and Volscians. 

T\Taen the truce was drawing to its close, the Veientines sent 
ambassadors to all the other Etruscan towns to solicit their 
assistance against the Romans; but it was refused, because 
another and far more dangerous enemy had appeared in the 
Apennines, and after the fashion of a Turkish invading army 
destroyed everything that came in their way : these were the 

' Two wars are here related, but they are, according to all appeai-ance, trans- 
posed ; the minute account of one at least is out of place, and probably belongs 
to the year 329, although hostilities may have occun-ed in 320 also; this at all 
events is the Chronology of Diodorus, to ■\vliich we must adhere. — N. 

* The Emperor Augustus du-ected Livy's attention to the fact, that Cossus, on 
the groimd of having gained the spolia opima on tliat occasion, set liiiuself up as 
consul, for that on his ai-mour he called himself consul. This is a later addition 
in Livy, and stands quite apart from his naiTative, for otherwise he ought to 
have placed the event seven years later. — N. 


Gauls. The Etruscans advised the Veientines by all means to 
maintain peace with the Romans; but the demands of the latter 
may have been very high; they may have claimed the sove- 
reignty of Veii, so that the Veientines were compelled to decide 
upon war. If we compare tlie account of the first Veientine 
war which occurred seventy years before, we find the Veien- 
tines at that time supported by all the powers of Etruria, 
whereas now they were confined to the protection of the 
Capenatans^ and Faliscans; it w^as only in one campaign that 
the Tarquinians came to their assistance; the Caerites were on 
friendly terms with the Romans and therefore remained neu- 
tral: the Etruscans indeed were masters of the place^ but the 
population may have been still essentially Tyrrhenian. In 
short, the war was limited to the Veientines and their imme- 
diate neighboui's. Rome was obliged to make the greatest 
efforts, and was supported by the Latins and Hernicans. 


The ridicule which Florus casts upon the hella suburbana : De 
Verulis et Bovillis pudet dicere sed triumphavimus, is that of a 
rhetorician, and we cannot quarrel w4th him for finding those 
occurrences uninteresting. "Wars carried on in a limited terri- 
tory cannot indeed have the same interest, as, for example, the 
Hannibalian war, but still we may see in them the development 
of the strength of Rome. We will not despise this Veientine 
war, yet we shall not describe it as minutely as Livy does, but 
confine our account to a few brief outlines. The feelings and 
sentiments with which the Romans undertook it deserve our 
admiration, for, considering their circumstances, the difficulties 
were as great as those which they had to encounter at the be- 
ginning of the first Punic war; it was only by continued 
perseverance that they could hope for a favorable issue. A city 
situated at so short a distance, and so well fortified as Veii 

* The Etniscan town of Capena was probably as near to Rome as Veii, 
tliotigh its site cannot be determined, because it disappears from history at an 
early time; hut it was ccrtaiidy situate between Veii, Falorii. and the Tiber. — N. 


could be conquered only by a blockade or siege ; for whenever 
the Veientines felt that they were too weak in the open field, 
they retreated within their walls, against which the Romans 
were powerless. It was necessary, therefore, either to blockade 
the town so as to compel it to surrender by hunger, and if 
necessary by fortifications and undermining, or the Romans 
had to try to reach it by inflicting calamities upon it, that is, 
they might fortify a place in the neighbourhood (eTTirei^^io-i?), 
as Decelea was fortified in the neighbourhood of Athens, and 
thence destroy everything far and wide, preventing all culti- 
vation of the fields, so that the hostile city would be thrown 
into a state of distress, which it must endeavour to avoid in 
every possible way. But in order to do this, the Romans 
would have been obliged to chano-e their mode of warfare, and 
moreover they had to fear the neighbouring towns of Capena 
and Falerii. Hitherto they had only made short campaigns 
during a few summer months, which often lasted only from 
ten to twelve, nay, sometimes not longer than six days, espe- 
cially during the time of the republic, for under the kings i 
must have been difierent. There were from the earliest times 
certain months destined for war, durino; which neio'hbourin^ 
tribes ravaged each other's territories: such was the case among 
the Greeks, and such is still the practice of the Asiatics. 
Russia and Persia fight every year for a few months on the 
frontiers of Georgia; and in the laws of Charlemagne the time 
is fixed during which nations are bound to serve in the field. 
In the intervals, intercourse between the coimtries was more 
or less free, and at the times of festivals especially it was quite 
free, as, e. g. during the common festivals of the Etruscans at 
the temple of Voltumna, and of the Ausonian nations at the 
temple of Feronia. The soldiers could be kept in the field for 
a limited time only, and when that was over they dispersed. 

The means of Rome for keeping up a great army had been 
much reduced since the Etruscan and Volscian wars. In former 
times, the armies had been paid out of the tithes which were 
paid by the possessor of the ager publicus, but since the ager 
publicus had been lost, every one went into the field as an 
olK6(rLTo<i, that is, he brought his provisions with him from 
home, and whatever more was required, he obtained by forag- 
ing; if this could not be done, the army was obliged to return 
home. Hence we hear so little of sieges. But now when the 

VOL. I. R 


war was to be conducted seriously, and as arms were not to be 
laid down till Veii sliould be subdued, it became necessary to 
pay tlie soldiers. This determination was perhaps connected 
with the proposal to recommence levying the tithes upon the 
ager publicvis, and to pay the soldiers out of their produce. 
But what seems to confirm the supposition that a stipendium 
was generally paid even from the earliest times, is the state- 
ment that in the census of Servius Tullius, the equites received 
2000 asses : if so, why should not the pedites also have received 
something? I conjecture that they were paid 100 asses, whe- 
ther the war lasted a longer or a shorter period : and out of this 
sum the soldier had to provide himself with arms and provi- 
sions. So long as this was the case, wars of conquest were 
impossible, for in them the soldiers must be entirely maintained 
by the state, and this latter is the arrangement which was made 
when, according to tradition, the Koman soldiers first received 
a stipendium. It would be wrong to suppose that before that 
time they had no pay at all ; but the diiference between receiv- 
ing a small sum once for all, or a small daily pay is considerable. 
We may take it for granted, that the aerarli, being exempt 
from military service, were at all times obliged to pay a war- 
tax for the pedites, as the o?-bi orhaeque did for the equites; for 
it is impossible that the double burden of serving with his life 
and his property should have fallen on the plebeian. 

The pay of the Eomans then, from early times, was 100 
asses per month for every man, and this pay was proportioned 
to their wants. Such pay was invariably given at Athens 
after the time of Pericles, but probably not earlier. The pay 
of an Athenian hoplite was enormous, but at Rome, where 
the allies did not pay any contributions, it was necessarily 
smaller. One hundred asses continued in later times to be 
the monthly pay. When the asses were made too lights they 
were calculated in silver in the proportion of one to ten. 
Every three days the soldier received a denarius (the value of 
a drachma), that is, daily two oboli. The stipendium was 
regarded as a unit, but was afterwards multiplied (multiplex 
stipendium. Domitian added a quartum stipeiidium). This, 
however, is at all times to be understood only of one month. 
The excellent Radbod Hermann Scheie errs in supposing, on 
the authority of writers who are worth nothing, that the 
stipendia were annual, which is in itself impossible, and would 


have answered no purpose. In this supposition, he was for 
once deserted by his practical good sense. The pay was 
given only for the time during which the soldiers were 
actually in the field. If the war lasted a whole year, pay was 
of course given for a year. Livy, in making Appius Claudius 
say, annua aera habes, annuam oj)eram ede, merely makes him 
express his own erroneous opinion. 

This innovation was of extraordinary importance to the 
republic; for without a national army, Rome would never 
have become great. Now if it had been possible to give the 
pay without imposing a tax, it would have been fortunate 
indeed; but if the patricians did not pay the tithes on the 
ager publicus, or if the revenue of the state was not sufficient, 
the plebeians must have felt the war very oppressive, for it 
then became necessary to obtain the pay by a property-tax, 
and it might so happen that the service in the army lasted a 
very long time. This injustice, however, was necessary. The 
plebeians had formerly not been taxed, probably because they 
had not been able to pay; but during the last twenty years, 
Rome had been ever increasing in prosperity, so that it Avas 
now possible to tax the plebeians, although it gave rise to 
new distress, which led again to the old oppression exercised 
by creditors on debtors. An army, however, might now be 
kept in the field all the year round. At the same time a 
change was made in the armour. Livy says, posfquam stipen- 
diariifacti sunt, scuta pro clupeis hahehant ; seeming to suppose 
that this change was occasioned by the introduction of pay for 
the soldiers. The first step to it may certainly have been 
taken even previously to the Gallic wars. 

The Romans began the last Veientine war with the deter- 
mination to conquer Veii. The republic, which now again 
extended as far as Anxur, began to feel her own strength, 
since she had gained the victory over the Aequians, and was 
at least on friendly terms with the Sabines. How far the 
Latins took part in this war is uncertain, but it is likely that 
their co-operation did not extend beyond the Tiber. It is not 
an improbable statement, that soon after Anxur, Circeii also 
was recovered by the Romans; so that the only place which 
still maintained its independence as a Volscian state was 
Privernum, a town at the foot of the hills. The weakness 
of the Ausonian nations was the result of the extension of the 


Samnites, and must have inclined tlaose nations to keep peace 
with Rome. The Romans thus had time to make a per- 
manent conquest of countries which probably they had no 
lonirer to share with the Latins. 

The preceding war with the Veientines had been succeeded 
by a truce of twenty years. The Etruscans, like many other 
nations of antiquity, were accustomed to conclude their wars 
by a mere truce for a certain number of years, and these years 
consisted of ten months. A proof of the latter assertion is the 
fact, that in nearly all instances, hostilities break out sooner 
than could have been expected if the years had been years 
of twelve months, but never earlier than would have been the 
case if the years were reckoned equal to ten months. The 
truce between Rome and Veil had been concluded in A.U. 330, 
and is said to have ended {induciae exierunt) in A.U. 347. The 
use of this year of ten months was very common among the 
Romans: it was the term established for mourning and for all 
money transactions. In the sale of corn, ten months' credit 
was a matter of course. Loans for a niunber of years were 
unknown, and all business was done only for short periods, 
being founded on personal credit like debts arising from bills 
of exchange. The Veientines, contrary to their former prac- 
tice, endeavoured to avoid the war by every possible means. 
There can be no doubt, that probably in consequence of its 
situation, Veii had formerly been the head of many Etruscan 
towns; for in previous wars its power appears very great. 
But the invasion of the Gauls caused the towns south of the 
Apennines, such as Arretium, Faesulae, and others, to be 
called upon to assist their countrymen beyond the mountains. 
This assistance was useless; the loss was great, and Etruria 
wasted her life-blood in the plains of Lombardy. Tarquinii 
and Capena alone supported Veii; the Aequi Falisci did the 
same, though not as an Etruscan people, but because they 
looked upon Veii as their bulwark. 

The Romans at first believed they could bring the war to a 
speedy termination. They built strong forts in the Veientine 
territory, just as Agis did in Attica during the latter period of 
ths Peloponnesian war ; and issuing from those forts, they 
prevented the Veientines from cultivating their fields, or 
burnt the ripe crops, so that distress and famine soon shewed 
themselves in the city. This system of warfare is designated 


on this occasion by the word obsessiu. The Eomans only once 
undertook to carry on a siege according to the simple manner 
of the time. Between two forts, a line parallel to the wall of 
the city was formed, consisting of accumulated rubbish, sacks 
of sand and fao-oots, and on both sides of it wooden fences 
(^plutei) were erected to give consistency to the rampart. It 
was pushed nearer and nearer to the city, an operation which 
presented the main difficult}'. These works were raised to 
about the same height as the wall, to which bridges and 
scaling ladders were applied {aggerem muru injungebant), and 
then the engines were brought up; at first battering-rams, but 
in later times catapulta and ballistae, for these engines, which 
were not yet known at Kome, were invented at Syracuse for 
Dionysius. The besieged endeavoured to destroy the works 
of the besiegers by imdermining them. But the neighbouring 
tribes defeated the Romans, and destroyed their Avorks; and 
from that moment several years passed away without the 
Romans again pitching their camp at Veii. 

The war of Veii presented to the ancients a parallel to the 
Trojan war; the siege was believed to have lasted ten years; 
and the taking of the city was as marvellous as that of Troy 
by the wooden horse. But the account of the whole war is 
not fictitious: the ancient songs took up isolated historical 
points, which they worked out and embellished, and this con- 
stitutes the difference between them and the lays of the earliest 
history. An annalistic narrative which is by no means in- 
credible, runs parallel to these lays. The defeat of the tribunes 
Virginius and Seririus is historical, but the detail about the 
Alban lake and the like belong to poetical tradition, and must 
be taken as the ancients give them : whether they were com- 
posed in prose or in verse is a matter of no consequence. The 
story runs as follows : — 

After Rome had already worn herself out in the struggle 
with Veii for upwards of seven years, and in the midst of 
the most profound peace with the Aequians and Volscians, 
a prodigy appeared. The Alban lake, the waters of which 
had always been below the edge of the ancient crater, began 
to rise, and threatened to overflow the country: this is 
the general substance of the ancient story, for in regard to 
the detail the accounts contradict one another; according to 
Dionysius and Dion Cassius (in Zonaras), a stream of water 


flowed from the lake straight towards the sea, while ac- 
cording to others, the lake only threatened to overflow 
its borders. The Romans knew not what to do; they had 
fixed their posts near Veii; whenever there was no fighting, . 
they observed a kind of truce : on one of these occasions, an 
Etruscan aruspex ridiculed the Romans for taking so much 
trouble to make themselves masters of Veii: so long as they 
were not masters of the Alban lake, said he, they could not 
take Veii. One Roman took notice of this remai"k, and 
under the pretext of a procuratio rei domesticae invited the 
aruspex to the camp. When he came, the Romans arrested 
him, and compelled him to say what was to be done. He 
answered that they must let off the waters of the Alban 
lake, so that they might be conducted through one of the 
nei2:hbourin2: rivers to the sea. The same answer was 
given by the god of Delphi. The Romans now undertook 
the work and finished it. AVhen it was nearly completed, 
the Veientines sent an embassy to entreat the Romans to 
receive their city in deditionem ; but the Romans would 
not listen to the proposal, for they knew that the talisman 
was broken. The Veientines did not contradict this, but 
said that it was also written in their books that if Veii 
should be destroyed, Rome would likewise soon be taken 
by barbarians, and that this part of the prophecy had been 
concealed from them by the aruspex. The Romans deter- 
mined to run the risk, and appointed Camillus dictator; he 
called upon all the people to take a part in securing the 
booty, and undertook to storm the city: all duties towards 
the gods were discharged, and human prudence now did 
its work. Camillus formed a subterraneous passage which 
led to a spot under the arx of Veii, and from that point a 
passage was made to the temple of Juno; for fate had de- 
termined that whoever should offer up the sacrifice in the 
arx of Veii, should win the victory. The Romans penetrated 
into the temple, slew the Etruscan king, and offered up the 
sacrifice. At this moment the walls of Veii were scaled by 
the Romans on all sides. 

This is very pretty poetry, and if we examine the historical 
nonsense of this account, we cannot hesitate for a moment 
to believe in the existence of a poem. The arx of W'ii is 


still discernible': it is situated near tlie aqua rossa, almost 
entirely surrounded by water, and is of a moderate height: 
the rock consists of tufo. The Romans would have been 
obliged to make their tunnel under the river, and to construct 
it so cunningly that no one should perceive it, and that at last 
they would only just have had to lift up the last stone in the 
temple and to rise above ground as through a trap-door. 

The fact was probably this. There were two kinds of sieges : 
the first was the one described above, which consisted in throw- 
ing up a mound of earth against the wall: according to the 
second method, the walls Avere undermined with immense 
labour, so that they rested only on a scafiblding of beams, 
which was then set on fire, and thus the downfall of the walls 
was caused. Battering rams probably do not occur previously 
to the Peloponnesian war, and among the Romans not till a 
still later time. If Veii was actually taken by a cuniculus, it 
must be understood to have been by undermining the walls. 

The letting off of the water of the Alban lake must certainly 
belong to this period; for there is no reason for supposing that 
a work of an earlier date was inserted here. The subterraneous 
passages through whicli the water was carried off had probably 
become obstructed, and Latium was in danger of being in- 
undated; it is possible that use was made of the credulity of 
the people to induce them to undertake this gigantic work, 
but I believe that when the senate declared the work to be 
necessary, there was no want of obedience. It must be supposed 
that the Alban lake, like the Fucinus and all other lakes formed 
in the craters of volcanoes, discharged its waters by subter- 
raneous passages, which may have been filled up in consequence 
of earthquakes. Livy, at a somewhat later time, speaks of a 
severe winter during which the Tiber was covered with ice, 
and says that epidemics were very prevalent in the following 
summer: the newly discovered excerpts from Dionysius place 
the building of the tunnel in the year alter that severe winter. 
Livy relates, that during that winter the snow was seven feet 
deep and that the trees were destroyed by frost; a statement 
closely resembling the records of the annals, and quite credible, 
although the ancient annals perished in the conflagration, for 
such a winter must have been remembered by every one. 

' Ii is a mere accident that I never was at Veii, but I have an acciu-ate 
knowledge of the locality from maps and drawings. — N. 


The winter of'A. U. 483 was equally severe, for suow covered 
the forum for forty days. The early history of Rome would 
indicate that the mean temperature of the air was then much 
lower than it is now.^ In the history of Rome and Greece, 
such unusual phenomena in the weather are nearly always 
followed by fearful earthquakes : an eruption of Aetna occurred 
about that time (a. u. 354). Vesuvius was then at rest, but 
the earthquakes were fearful, and it is possible that by one of 
them the outlets of the Alban lake were obstructed. In 
general all lakes which have no outlet, have remarkable periods 
in the increase and decrease of their waters. Lake Copa'is had 
even artificial tunnels to let off the water into the sea of Euboea : 
these were afterwards obstructed, and in the time of the 
Macedonians all Boeotia was not able to raise the money neces- 
sary for cleansing them ; in consequence of which the waters 
rose and inundated the country far and wide. After that time 
probably nothing was done to remedy the evil; it is very 
likely, as Aristotle observes, that the quantity of water in 
Greece was decreasing ; lake Copais is at present a mere swamp 
and in reality no longer deserves the name of lake, but re- 
sembles the bogs in our moors. 

What the Romans did to prevent the overflow of the 
water is extraordinary : the tunnel still exists uninjured of the 
length of 2,700 paces or nearly three Roman miles ^, and the 
water of the lake is reduced to a proper height. The advan- 
tages of letting off the water are great even now, although 
the coimtry around is uncultivated and covered with shrubs 
and bushes, since it supplies water for domestic purposes to the 
( .ampagna of Rome : and although this water is not very good, 
still it is better than that in the wells of the Campagna. The 
tunnel resembles the greatest Etruscan works: the entrance 

■■' Conipiirc, on the uthcr hiuul, Hist of Jiume, vol. iii. n. 1034. I was unwill- 
i ng to supjiress tlic above jjussage occnrring in the Lectures of 1826-7. According 
to Arngo, the winter in Tuscany is less cold find tlie snmmer less hot than 
tbrnierly. See Berglians, Lander mid Vblker. i. p. 248. — Eu. 

' This statement l)elongs to tlie Lectures of 1828-9, and is the same in all the 
manuscripts, but in 1826-7, Niebuhr said that the length of the tunnel had not 
hen measured, Imt tliat it wiis estimated at 7,500 feet ux two and a half Roman 
miles. In his Hist, of Home, ii p. 508, the length is said to be 6000 feet. 
Abekcn (M/«e/i<a/(c«, p.l79) says: "The tunnel runs into the s^outh-western 
side of the lake and is nearly 4000 feet long." The measurement given in the 
text seems therefore to be based upon an error.— Eu. 


from the lake into it is like the vault of a temple executed in 
the grandest style, whence we see that the Romans now again 
built in the same manner as under the kings : this is charac- 
teristic of the age of Camillus. The tunnel is cut for the most 
part through a hard mass of lava, only a small portion running 
through peperino which is more easily worked, and forms a pas- 
sage nine palms high and five palms broad. By this work, which 
has never yet required reparation, the lake, it seems, is per- 
manently confined to a limited height : it was previously about 
100 feet higher than the level to which it was then reduced. 
It is an interesting point to know how such a work was exe- 
cuted. Considering the imperfection of instruments in those 
times when the compass was yet unknown, it must have been 
extremely difficult to find the correct point at a distance of 
more than two miles; and even now it would be attended with 
great difficulties, for the architect must know to a line how 
high he may build in order to find the inclined plane for the 
watercourse. It is well known in the country and is also re- 
corded in some books, that on the whole line from the lake to 
the point in the plain whither the water was to be conducted, 
there exist to this day open wells into which people descend to 
cleanse the tunnel ; but these openings were not made merely 
to enable the mud to be removed — the lake is not muddy — 
but also to calculate the depth and let in air. By means of 
these wells, the Romans were enabled accurately to calculate 
the line as far as the issue. In our days people have so little 
practice in levelling, that till very recently it was not known 
that the lake of Nemi lies higher than that of Alba.-* The 
construction of these wells rendered it possible to employ a 
great many men at once, and to complete the work with toler- 
able speed : from the bottom of each well two parties of work- 
men might work in opposite directions and so as to meet other 
parties commencing at the bottom of other wells. In this 
manner the tunnel was formed till it came close up to the lake, 
the entrance to which was undoubtedly made by means of a 
stone-borer of the thickness of a little finger, since the wall of 
basalt need not have been thicker than two cubits to resist the 
whole pressure of the lake. The small opening being made, 
the water ran off very gradually, so that the workmen had 
time to be pidled up through the wells. W^ien this was done, 
* It must be remembered that this was said in the year 1828. 


the wall between the lake and the tunnel was knocked down, 
and the entrance facing the lake was made in such a manner 
as to prevent trees and the like from being carried into the 
tunnel; tlie arch was then embellished and wrought into a 
magnificent portico, like the entrance to a temple. Tliis struc- 
ture eclipses all the works of Egypt: they are wonderful but 
useless; this is practical and useful. 


That Veil was taken by storm is certain: the people were 
destroyed, and the city was methodically plundered. It is 
related, that the whole Roman population was let loose upon 
the place for the purpose of plundering it ; but this can refer 
only to the men capable of bearing arms, and was done partly 
on account of the proximity of Rome, and partly because in 
the long protracted war all had taken a part in it. The fate 
of the inhabitants of Veii was the same as that which befell so 
many people of antiquity: whoever escaped from the sword, 
was led away into servitude. When the city fell into the 
hands of the Romans, it was empty. We readily believe that 
it was a more beautiful place than Rome. The latter has an 
excellent situation, but its picturesqueness is connected with 
many disadvantages, for the territory round the city is exposed 
to frequent inundations, and the intercourse within was very 
difficult for carriages and other conveyances, on account of the 
many hills and valleys. Veii, on the other hand, with the 
exception of its arx, was situated in a plain , and probably had 
beautiful and wide streets: no wonder therefore that the 
Romans were loth to destroy the handsome city. 

Immediately after the conquest, quarrels arose between the 
gtjvernment and the plebcs who demanded a distribution of 
the territory, while the former claimed the whole for tliem- 
selves alone; but such a thing was no longer possible. Another 
difficulty was occasioned by the beauty of Veii, for it was 
thought a pity that it should remain deserted: it may easily 
be conceived that Avhcn it was proposed to distribute the ter- 


ritory, It was at the same time wished that those who were in 
want of houses should have habitations assigned to them at 
Veii. A tribune of the people proposed that if the patricians 
thought the plebeians too low to live together with them, the 
plebes with their magistrates should emigrate to Veii: it would 
be folly to believe that the proposal was such as Livy describes 
it, viz. : that half of the senate and the people should remove 
to Veil. But even the other proposal calls forth doubts, for 
the scheme would have been highly unreasonable, and the 
objections which Livy raises against such a tearing up of the 
republic are very important, and after all, a complete separa- 
tion would have been impossible. It would have been dan- 
gerous even to discuss the sending of a great colony with a 
local government to Veii. But an arrangement was made: 
the patricians secured the greater part of the territory for their 
occupation, but the plebes also obtained a portion, and not only 
each for himself the usual seven jugera forensia, but also 
something in consideration of his children. According to 
Diodorus, each flimily received twenty-eightyw^era; but if this 
is true, the territory of Veii must have been enormous. The 
aerarii had no share in these assignments^ but those among 
them who were clients of patricians received allotments on the 
farms of their patrons. 

The sequel of these events shoAvs, that at that time the ter- 
ritory of Veii and Capena, and of the Etruscan cities in 
general, comprised large tracts of country with subject towns, 
which, during the war threw themselves into the arms of the 
Romans: those who did so were undoubtedly the ancient 
inhabitants of those places, who saw in the Romans their 

The conquest of Veii was one of the most decisive events in 
history, for it delivered Rome from a counterpoise which checked 
her development. As all the east of Latium was at peace, the 
Romans irresistibly penetrated into Etruria, the Etruscans being 
obliged in the mean time to exert all their powers in the Apen- 
nines to keep oif the Gauls. But the war was also carried on 
against the Faliscans: to judge from their name they were 
A^oliscans, whence Virgil calls them Aequi Falisci. According 
to Strabo, they had a peculiar language, and were a nation 
different from the Etruscans. The war of Camillus against 
them is known to us all from our childhood; the talc, that by 


his generosity he influenced them so much, that they accepted 
tlie treaty of friendship with the Romans, has great internal 
improbability: the story of the schoolmaster's treachery I may 
leave uncriticised. The war was moreover directed against the 
Vulsinians, in whose territory the Romans made conquests and 
with whom they concluded an advantageous peace. The Ro- 
mans had then penetrated even beyond the Ciminian forest, 
which afterwards in the great war of Fabius presents so fright- 
ful an ajjpearance. The separation of countries by this forest 
does not seem to have been very strongly marked at that time, 
but afterwards it appears to have been intentionally allowed to 
grow wild, that it might form a boundary, just as a forest 
divides the Austrian from the Turkish part of Dalmatia. Ca- 
pena is not afterwards mentioned in history, and must therefore 
have been destroyed either by the Romans after the taking of 
Veii or by the Gauls : it is an historical fact, that subsequently 
to the Gallic calamity the surviving Capenatans became Roman 

After these victories, Camillus shone as the greatest general 
of his age; but he was nevertheless accused of having appro- 
priated to himself sundry valuable things from the Veientine 
booty, especially the brass gates of the temple of Jimo, and of 
having declared too late his obligation to dedicate the tenth 
part of the booty to the Pythian Apollo. It would be an un- 
profitable labour to speculate on the guilt or innocence of 
Camillus; but we must not forget that every Roman general 
was entitled to set apart a portion of the booty to himself We 
cannot decide whether Camillus took more than he was legally 
allowed or not; what one person does on a small scale another 
does on a large. We must not believe that Camillus committed 
the crime in secret, for he undoubtedly ordered the gates to be 
fitted to his own house; if he had wished to make use of the 
metal they would have been melted down long before. The 
real cause of the hatred against Camillus was of a political kind, 
for down to the end of his life he stood at the head of the most 
stubborn patrician party; the plebeians were ever becoming- 
stronger and more powerful, and the ease of prosperity had 
produced a certain desire for excitement : Camillus was accused 
because a considerable party was against him, and he was sen- 
tenced to pay a sum of 15,000, or according to others 100,000 
' Compare the speech of Fuhricius in Dionysms (p. 77, cd. Sylb.) — N. 


or even 500,000 asses. He did not wait for the sentence but 
went into exile to Ardea. Livy says that previously to the 
trial he implored his clients and tribules to do their utmost to 
obtain his acquittal (this would prove that he was tried before 
the centuries, for the tribes cannot possibly be meant) : they 
are said to have declared that they would pay his fine, but not 
acquit him : this would clearly prove his guilt. According to 
Dionysius, his gentiles and clients actually paid the fine and he 
went into exile from vexation. I believe however that he was 
condemned by the curies, because when he was to be recalled, 
the curies assembled in the Capitol to repeal the decree of ban- 
ishment; for the curies could assemble only at Rome, and this 
would prove likewise that he was found guilty, — a thing then 
not uncommon with great men. 

At that time no Eoman foresaw the calamity which was 
threatening them. Rome had become great, because the 
country which she had conquered was weak through its oli- 
garchical institutions; the subjects of the other states gladly 
joined the Romans, because under them their lot was more 
favorable, and probably because they were kindred nations. 
But matters went with the Romans as they did with Basilius, 
who subdued the Armenians when they were threatened by 
the Turks, and who soon after attacked the whole Greek em- 
pire and took away far more than had been gained before. 

The expedition of the Gauls into Italy must be regarded as 
a migration, and not as an invasion for the purpose of conquest : 
as for the historical account of it, we must adhere to Polybius 
and Diodorus, who place it shortly before the taking of Rome 
by the Gauls. We can attach no importance to the statement 
of Livy that they had come into Italy as early as the time of 
Tarquinius Prisons, having been driven from their country by 
a famine : it undoubtedly arose from the fact, that some Greek 
writer, perhaps Timaeus, connected this migration with the 
settlement of the Phocaeans at IMassilia. It is possible that 
Livy even here made use of Dionysius, and that the latter fol- 
lowed Timaeus: for as Livy made use of Dionysius in the eighth 
book, why not also in the fifth? He himself knew very little 
of Greek history-; but Justin's account is here evidently op- 
posed to Livy. Trogus Pompeius was born in the neighboxar- 
hood of Massilia, and in writing his forty-third book, he 
' Comp. Hist of Borne, vol. iii. n. 485. 


obviously made use of native clironicles, for from no other 
source could he derive the account of the decreta honorifica of 
the Romans to the Massilians for the friendship which the latter 
had shown to the Romans during the Gallic war; and from 
the same source must he have obtained his information about 
the maritime wars of Massilia against Carthage. Trogus knows 
nothing of the story that the Gauls assisted the Phocaeans on 
their arrival; but according to him, they met with a kind 
reception among the Ligurians, who continued to inhabit those 
parts for a long time after. About the year a.u. 350, that is, 
about fifteen years before, Livy himself says, gentem invisiiatem^ 
novos acculas, Gallos comparuisse. Even the story of the Lucumo 
who is said to have invited the Gauls is opposed to him, and if 
it were referred to Clusium alone, it would be absurd. Poly bins 
places the passage of the Gauls across the Alps about ten or 
twenty years before the taking of Rome ; and Diodorus describes 
them as advancing towards Rome by an uninterrupted march. 
It is further stated, that Melpum in the country of the Insu- 
brians was destroyed on the same day as Veii : without admittino- 
tlois coincidence, we have no reason to doubt that the state- 
ment is substantially true ; and it is made by Cornehus Xepos 
who, as a native of Gallia Transpadana, might possess accurate 
information, and whose chronological accounts were highly 
esteemed by the Romans. There was no other passage for the 
Gauls except either across the Little St. Bernard or across the 
Simplon: it is not probable that they took the former road, 
because their country extended only as far as the Ticinus, and 
if they had come across the Little St. Bernard, they would 
naturally have occupied also all the country between that 
mountain and the Ticinus. The Salassi may indeed have been 
a Gallic people, but it is by no means certain ; moreover be- 
tween them and the Gauls who had come across the Alps the 
Laevi also lived ; and there can be no doubt that at that time 
Ligurians still continued to dwell on the Ticinus. 

Melpum must have been situated in the district of INIilan. 
The latter place has an uncommonly happy situation : often as 
it has been destroyed, it has always been restored, so that it is 
not impossible that IMelpum may have been situated on the 
very spot afterwards occupied by IMilan. The Gallic migration 
undoubtedly passeil by like a torrent with irresistible rapidity: 
how then is it possible to suppose that I\lelpum resisted them 


for two centuries, or that tliey conquered it and yet did not 
disturb the Etruscans for two hundred years? It would be 
absurd to believe it, merely to save an uncritical expression of" 
Livy. According to the common chronology, the Triballi who 
in the time of Herodotus inhabited the plains, and were after- 
wards expelled by the Gauls, appeared in Thrace twelve years 
after the taking of Kome (according to a more correct chrono- 
logy it was only nine years after that event). It was the same 
movement assuredly which led the Gauls to the countries 
through which the middle course of the Danube extends, and 
to the Po : and could the people who came in a few days from 
Clusium to Eome, and afterwards appeared in Apulia, have 
been sitting quiet in a corner of Italy for two hundred years? 
If they had remained there because they had not the power to 
advance, they would have been cut to pieces by the Etruscans. 
We must therefore look upon it as an established fact, that the 
migration took place at the late period mentioned by Polybius 
and Diodorus. 

These Gauls were partly Celts, and partly (indeed prin- 
cipally) Belgae or Cymri, as may be perceived from the 
circumstance, that their king, as well as the one who appeared 
before Delphi, is called Brennus. Brenin, according to Ade- 
lung, in his Mithridates, signifies in the language of Wales 
and Lower Brittany, a king. But what caused this whole 
emigration? The statement of Livy, that the Gauls were 
compelled by famine to leave their country, is quite in keep- 
ing with the nature of all traditions about migrations, such as 
we find them in Saxo Grammaticus, in Paul Warnefried from 
the sagas of the Swedes, in the Tyrrhenian traditions of 
Lydia, and others. However, in the case of a people like the 
Celts, every specific statement of this kind, in which even 
the names of their leaders are mentioned, is of no more value 
than the traditions of other barbarous nations which were a 
unacquainted with the art of writing. It is indeed well 
known that the Celts in writing used the Greek alphabet, but 
they probably employed it only in the transactions of daily 
life; for we know that they were not allowed to commit their 
ancient songs to writing. There was, however, among the 
Celts a tradition which we find in Ammianus ]\Iarccllinus, 
that Britain was one of their earliest seats. Now we meet 
with them in different parts of Britain, Ireland, and Spain, 


and in two places of Portugal, for the Celtic! and Celtae in 
Portugal, wlio lived in Algarbia and Alemtejo, and between 
the j\Iinho and Douro, were pure Celts. The Celtiberians in 
Spain were a mixture of Celts and Iberians, and dwelt in the 
very heart of the mountains between Saragossa and Madrid, 
which are connected with the Pyrenees. ^ There was the 
same tradition about those Celts in Spain as about their 
appearance in Italy, for they were said to have been driven 
thither by a famine, and then to have made conquests and 
spread over the country. Here again we have a confusion of 
the polarity in history. Wlaerever there is a national migra- 
tion, we never find the invading people settled in scattered 
groups. The dispersed inhabitants of such countries, especially 
in mountainous districts, are usually remnants of the ancient 
population, the bulk of which has either emigrated or become 
changed. Among the Celtiberians the Iberians predominated, 
but the Celts were the native people, united with the Iberians 
who immigrated from Africa. The language which arose 
from this union may have been a mixture of the two, but the 
names of places are IlDerian. Similar changes of a great tribe 
do sometunes occur in history. The Wends in Germany, 
who were originally Slavonians, though the colonies founded 
among them were not of much consequence, and there was 
neither a German conquest, nor German princes, yet for the 
most part adopted the German language. 


The existence of the Aipiitanians in Gaul, is a proof of the 
mif^ration of the Iberians, for Ave are told by Caesar that they 
were pure Spaniards; and there is no reason for supposing 
that this migration took place at a late period, for the 
Basques still live north of the Pyrenees. We have, moreover, 
the statement of Scylax, that the people of Gaul, from the 
Pyrenees to the river Rhodanus, was composed of Iberians 
and Ligurians. The Celts at one time were masters of all 
* The mountains of southern Spain are connected with those of Africa. — N. 


Spain, with the exception of Andalusia; and besides Spain, 
they occupied the south of France, Ireland, and a part of 
England. The boundary of the Iberians cannot be accurately 
fixed in the north, though in the earlier times it was the 
Sierra Morena. In the south we find them in southern 
Spain, in the Balearean islands, Sardinia, Corsica, and West- 
ern Sicily; and lastly also in Africa. 

The Cymri or Belgae were a people different from the 
Celts, though akin to them. This diflference, concerning 
which I expressed my opinion years ago, is of great import- 
ance, and is now generally acknowledged. It is not a new 
discovery; I have only brought forward facts which were 
previously overlooked. Caesar's idea that the Belgae were a 
mixture of Germans and Celts is erroneous. They were per- 
fectly distinct from the Germans, although a small number of 
words in their language are Germanic. In Caesar's time they 
were unquestionably Cymri, somewhat mixed with Germans, 
who had joined them in their migration. A part of Britain 
too was inhabited by Cymri, who were probably the earlier 
inhabitants, having afterwards been expelled by the Gael. 
The latter were pressed on by the Iberians, the Cymri by the 
Gael, and the Germans by the Cymri, who then inhabited the 
north of France and the low countries, which were subse- 
quently inhabited by Celts. 

The south of France from the Pyrenees, Lower Languedoc, 
and the valley of the Ehone, Piedmont, and Lombardy, as far 
as the Etruscans, were occupied by the Ligurians, a great 
European nation. Scylax states, that Lower Languedoc had 
a mixed population of Iberians and Ligurians, and in later 
times, which cannot be chronologically determined, the Celts 
drove the Iberians from Spain as far as the Garonne, and the 
Iberians forced the Ligurians to retire as far as the district of 
Aix in Provence, an event which may be recognised from its 
consequences. By this impulse the Gauls and Cymri together 
were compelled to emigrate : some Cymri retreated before the 
Gauls and went away, but others joined them. The Gauls 
and Cymri were very different from one another, for their 
language and grammar are quite distinct. The two great 
migrations under Bellovesus and Sigovesus, which are men- 
tioned by Livy, must be regarded as true, although the leaders 
are perhaps nothing but personifications. The one directed 

VOL. I. S 


towards Italy, between the Alpine tribes of the Etruscans and 
the Lignrians, overran the Etruscan towns in the plain of 
Lombardy ; the other extended north of the Alps. The 
Raetians, Lepontians, Camunians, Stonians, and other Alpine 
tribes in the Tyrol and the Southern Alps, as far as Verona, 
maintained their ground like islands among the invading 
Gauls, who poured in around them like a sea; so that their 
situation reminds us of that of the three Celtic tribes in Spain. 
The misfration of which the Helvetii were a remnant, has 
been sufficiently explained in my Essay on the Scythians and 
Sarmatians.' It first appears about the Black Forest, where 
it rested for a while, and thence proceeded towards the middle 
Danube, Hungary, and Lower Slavonia. There the migratory 
hordes undertook the difficult conquest of the high mountains, 
and then spread over Macedonia, Thrace, and Bulgaria. 
They also advanced across the Danube as far as the Dnieper, 
but being repelled by the Sarmatians, they again threw them- 
selves upon Europe. This is the only known instance in which it 
is clear, that the torrent of a migrating people rolls on till it 
meets with insurmountable obstacles, and then returns with 
the same rapidity. At the time when Herodotus wrote (about 
A.U. 320), the people on the middle and lower Danube still 
lived undistiirbcd in their ancient seats. The Scythians 
inhabited Moldavia and Wallachia as far as Transylvania. 
Slavonia and Lower Hungary were inhabited by the Agathyi'si 
and Triballi. But nine years after the taking of Rome by the 
Gauls, the Triballi appear in the ncighboui'liood of Abdera in 
Thrace, and afterwards we find them on the southcra part of 
the Danube in Bulgaria. The Scythians, on the other hand, 
were confined to Bessarabia as early as the time of Philip ; and 
in the time of Alexander, the Gctae were in possession of 
Moldavia and Wallachia. The nation that brought about this 
change was the Gauls, and that in the same migration dm-ing 
which they poured in upon Italy. 

Scylax (Olymp. 106) was aware that there were Gauls at 
the head of the Adriatic, which was afterwards inhabited by 
the Carnians and Noricans; they Avere, according to him, a 
remnant of the Gallic migration, and a part of the Gauls who 
had advanced fiirther dwelt in Sirmium; thence they crossed 
the Danube under the name of the Bastarnians, and forced the 

' Fn tlie Kleirte Hist. Sr/irif/cn, Erstc Sammhtng, i>.,3.'52, etc. 


Getae to tlii'ow themselves into Hungary and Transylvania ; 
afterwards they spread in the Ukraine. From the important 
inscription of Olbia published by Kohler, we see that the 
Galatians, and along with them the vSciri (afterwards a German 
people), lived about the Dnieper, and this fact agrees well with 
the disappearance of the Scythians at that time. For there 
was also an eastern migration of the Sarmatians, a people 
whom Herodotus knew only beyond the Tanais; Scylax, who 
lived seventy years later, speaks of them as living on the 
western side of that river: in the inscription of Olbia they 
appear east of the Dnieper, and under Augustus they destroyed 
the Greek towns in Wallachia. This movement afterwards 
caused the migration of the Cymri or Cymbri, for the Cymri 
always took part in the migration of the Celts : among them 
were the Bastarnae, who lived in the south of Poland and in 
Dacia, and were expelled by the Sarmatians: J. von Miiller was 
the first who saw the truth of the statement of Posidonius, that 
the Cimbri did not come from Jiltland but from the East; he 
did not see however that they were originally Belgie, or, as the 
Greeks call them by a more general name, KeXjai. It is fool- 
ish to claim the victories of the Cimbri as having been gained 
by the German nation. 

These migrations extended in Germany as far as the river 
Mayn and the forest of Thliringia, nay even Bohemia was 
inhabited by Celts previously to the time of Caesar, and some 
of their tribes existed in that country as late as the time of 
Tacitus, for the Gothini still continued to speak Gallic; and 
the Noricans in Austria were of the Celtic race. The Kaetians 
were Etruscans, and the Vindelicians Liburnians. .The Helvetii 
conquered the greater part of Switzerland, but the country 
about the St. Gothard remained in the possession of the ancient 
inhabitants. The Gauls penetrated into Italy only through a 
limited part of the Alps, probably across the Simplon, and the 
Valais was the sole bond between the Gauls in Italy and their 
kinsmen north of the Alps. As far as Aosta the ancient in- 
habitants maintained themselves, for the Salassians, Taurinians 
and others were Ligurians, and the people at the foot of the 
St. Gothard were Etruscans. The Ligurians were a very war- 
like people and kept their ground on both sides of the Alps; 
the Allobroges, however, were pure Celts. Hence Gallia 
Cisalpina in our maps is much too large, and that even in 

S 2 


D'Anvllle's map. Piedmont formed no part of it; it com- 
prised only the Austrian territory of Milan, Bergamo and 
Brescia, Lombardy soutli of tlie Po as far as the Adriatic, and 
north of the Po to about the lake of Garda. Thus all the 
country occupied by them was in the plain, and this is another 
reason why their migration cannot have lasted as long as Livy 

During this Gallic migration we are again made aware how 
little we know of the history of Italy generally : our knowledge 
is limited to Rome, so that we are in the same predicament 
there, as if of all the historical authorities of the whole German 
empire wc had nothing but the annals of a single imperial city. 
According to Livy's account, it would seem as if the only object 
of the Gauls had been to march to Rome ; and yet this immigra- 
tion changed the whole aspect of Italy. After the Gauls had 
once crossed the Apennines, there was no further obstacle to 
prevent their marching to the south of Italy by any road they 
pleased; and it is in fact mentioned that they did proceed 
farther south. The Umbrians still inhabited the country on 
the lower Po, in the modern Romagna and Urbino, parts of 
which were occupied by Liburnians. Poljdnus says that many 
people there became tributary to the Gauls^ and that this was 
the case with the Umbrians is quite certain. 

The first historical appearance of the Gauls is at Clusium, 
whither a noble Clusine is said to have invited them for the 
piirpose of taking vengeance on his native city. Whether this 
account is true, however, must remain undecided, and if there 
is any truth in it, it is more probable that the offended Clusine 
went across the Apennines and fetched his avengers. Clusium 
has not been mentioned since the time of Porsena ; the fact of 
the Clusines soliciting the aid of Rome is a proof how little 
that northern city of Etruria was concerned about the fate of 
the southern towns, and makes us even suspect that it was 
allied with Rome; however, the danger was so srreat that all 
jealousy must have been suppressed. The natural road for the 
Gauls would have been along the Adriatic, then through the 
country of Umbrians who were tributary to them and already 
quite broken down, and thence through the Romagna across 
the Apennines. But tlie Apennines which separate Tuscany 
from the Romagna are very dillicult to cross, especially for 
sumplcr horses; as therefore tlie (iauls could not enter Etruria 


on that side, which the Etruscans had intentionally allowed to 
grow wild, and as they had been convinced of this in an un- 
successful attempt, they crossed the Apennines in the neigh- 
bourhood of Clusium, and appeared before that city. Clusium 
was the great bulwark of the valley of the Tiber ; and if it 
were taken, the roads along the Tiber and the Arno would be 
open, and the Gauls might reach Arezzo from the rear: the 
Romans therefore looked upon the fate of Clusium as decisive 
of their own. The Clusines sued for a treaty with the mighty 
city of Rome, and the Romans were wise enough readily to 
accept the offer : they sent ambassadors to the Gauls ordering 
them to withdraw. According to a very probable account, the 
Gauls had demanded of the Clusines a division of their terri- 
tory as the condition of peace, and not, as was customary with 
the Romans, as a tax upon a people already subdued: if 
this is correct, the Romans sent the embassy confiding in 
their own strength. But the Gauls scorned the ambassadors, 
and the latter, allowing themselves to be carried away by 
their warlike disposition, joined the Etruscans in a fight 
against the Gauls: this was probably only an insignificant 
and isolated engagement. Such is the account of Livy, 
who goes on to say, that the Gauls, as soon as they 
perceived this violation in the law of nations, gave the 
signal for a retreat, and having called upon the gods to avenge 
the wrong marched against Rome. This is evidently a mere 
fiction, for a barbarous nation like the Gauls cannot possibly 
have had such ideas, nor was there in reality any violation of 
the law of nations, as the Romans stood in no kind of connec- 
tion with the Gauls. But it was a natural feeling with the 
Romans to look upon the fall of their city as the consequence 
of a nefas, which no human power could resist. Roman 
vanity also is at work here, inasmuch as the Roman ambassa- 
dors are said to have so distinguished themselves, that they 
Avere recognized by the barbarians among the hosts of Etrus- 
cans. Now according to another tradition directly opposed to 
these statements, the Gauls sent to Rome to demand the surren- 
der of those ambassadors: as the senate Avas hesitating and left 
the decision to the people, the latter not only rejected the de- 
mand, but appointed the same ambassadors to the office of mili- 
tary tribunes, whereupon the Gauls with all their forces at once 
marched towards Rome. Livy here again spealcs of the pnjmUia 


as the people to whom the senate left the decision : this must 
have been the patricians only, for they alone had the right to 
decide upon the fate of the members of their own order. It is 
not fair to accuse the Komans on that occasion of dishonesty; 
but this account assuredly originated with later writers who 
transferred to barbarians the right belonging to a nation stand- 
ing in a legal relation to another. The statement that the 
three ambassadors, all of whom were Fabii, were appointed 
military tribunes, is not even the usual one, for there is another 
in Diodorus, who must here have used Roman authorities 
written in Greek, that is, Fabius ; since he calls the Caerites 
KalpioL and not ^AyvWaioc. He speaks of a single ambassador, 
who being a son of a military tribune fought against the Gauls. 
This is at least a sign how uncertain history yet is. The 
battle on the Alia was fought on the 1 6th of July : the military 
tribunes entered upon their office on the first of that month ; 
and the distance between Clusium and Rome is only three 
good days' marches. It'is impossible to restore the true history, 
but we can discern what is fabulous from what is really his- 

An innumerable host of Gauls now marched from Clusiimi 
towards Rome. For a long time the Gauls were most formid- 
able to the Romans, as well as to all other nations with whom 
they come in contact even as far east as the Ukraine; as to 
Rome, we see this as late as the Cisalpine war of the year A.U. 
527. Polybius and Diodorus are our best guides in seeking 
for information about the manners of the Gauls, for in the time 
of Caesar they had already become changed. In the descrip- 
tion of their persons we partly recognise the modern Gael, or 
the inhabitants of the Highlands of Scotland : huge bodies, 
blue eyes, bristly hair; even their dress and armour are 
those of the Highlanders, for they wore the checked and 
variegated tartans {sagula virgata, versicoloria); their arms 
consisted of the broad unpointed battle-sword, the same 
weapon as the claymore among the Highlanders. They had 
a vast number of horns, which were used in the Highlands for 
many centuries after, and threw themselves upon the enemy in 
immense and irregular masses with terrible fury, those standing 
behind impelling those stationed in front, whereby they became 
irresistible by the tactics of those times. The Romans ought 
to have used against them their phalanx and doubled it, until 


they were accustomed to this enemy and were enabled by 
their greater skill to repel them. If the Romans had been 
able to withstand their first shock, the Gauls would have easily 
been thrown into disorder and put to flight. The Gauls who 
were subsequently conquered by the Romans were the descend- 
ants of such as were born in Italy, and had lost much of their 
courage and strength. The Goths under Vitiges, not fifty 
years after the immigration of Theodoric into Italy, were 
cowards and unable to resist the 20,000 men of Belisarius: 
showing how easily barbarians degenerate in such climates. 
The Gauls moreover were terrible on account of their inhuman 
cruelty, for wherever they settled, the original towns and their 
inhabitants completely disappeared from the face of the earth. 
In their own country they had the feudal system and a priestly 
government: the Druids were their only rulers, who avenged 
the oppressed people on the lords, but in their turn became 
tyrants: all the people were in the condition of serfs, — a proof 
that the Gauls, in their own country too, were the conquerors 
who had subdued an earlier population. We always find 
mention of the wealth of the Gauls in gold, and yet France 
has no rivers that carry gold-sand, and the Pyrenees were then 
no longer in their possession : the gold must therefore have 
been obtained by barter. i\Iuch may be exaggeration ; and 
the fact of some noble individuals wearing gold chains was 
probably transferred by ancient poets to the whole nation, 
since popular poetry takes great liberty especially in such 

Pliny states that previously to the Gallic calamity, the 
census amounted to 150,000 persons, which probably refers 
only to men entitled to vote in the assemblies, and does not 
comprise women, children, slaves and strangers. If this is 
correct, the number of citizens was enormous; but it must 
not be supposed to include the inhabitants of the city only, 
the population of which was doubtless much pmallcr. The 
statement of Diodorus that all men were called to arms to 
resist the Gauls, and that the number amounted to 40,000, 
is by no means improbable: according to the testimony ol 
Polybius, Latins and Hernicans also were enlisted. Another 
account makes the Romans take the field against the Gauls 
with 24,000 men, that is, with four field-legions and four 
civic legions: the field-legions were formed only of plebeians, 


and served, according to the order of the classes, probably 
in maniples; the civic legions contained all those who be- 
longed neither to the patricians nor to the plebeians, that is, 
all the aerarii, proletarii, freedmen, and artizans who had never 
before faced an enemy. They were certainly not armed with 
the pilum, nor drawn up in maniples; but used pikes and were 
employed in phalanxes. Now as for the field-legions, each 
consisted half of Latins and half of Romans, there being in 
each maniple one century of Romans and one of Latins. There 
were at that time four legions, and as a legion, including the 
reserve troops, contained 3,000 men, the total is 12,000; now 
the account which mentions 24,000 men, must have presumed 
that there were four field-legions and four irregular civic ones. 
There would accordingly have been no more than 6,000 
plebeians, and, even if the legions were all made up of Romans, 
only 12^000; if in addition to these we take 12,000 irregular 
troops and 16,000 allies, the number of 40,000 would be 
completed. In this case, the population of Rome would not 
have l^een as large as that of Athens in the Peloponnesian war, 
and this is indeed very probable. The cavalry is not in- 
cluded in this calculation : but 40,000 must be taken as the 
maximum of the whole army. There seems to be no exag- 
geration in this statement, and the battle on the Alia, speaking 
generally is an historical event. It is surprising that the 
Romans did not appoint a dictator to command in the battle ; 
it cannot be said indeed that they regarded this war as 
an ordinary one, for in that case they would not have raised 
so great a force, but they cannot have comprehended the 
danger in all its greatness. New swarms continued to come 
across the Alps; the Senones also now appeared to seek 
habitations for themselves; they, like the Germans in after- 
times, demanded land, as they found the Insubrians, Boians 
and others already settled; the latter had taken up their 
abode in Umbria, but only until they should find a more 
extensive and suitable territory. 

The river Alia possesses no remarkable feature, and one 
might almost be inclined to believe that the aspect of the 
country in that district has changed. It is only by the 
distances being mentioned that we can determine the river 
called Alia. The ancients describe it as a river with hi-^di 
banks, but the modern river which must be identified with it. 


has no such banks. The name has entirely disappeared. In 
summer all the rivers of that country have very little water, 
and the position behind it was therefore of little avail. The 
Romans committed the great mistake of fighting with their 
hurriedly collected troops a battle against an enemy who had 
hitherto been invincible. The hills along which the right 
wing is said to have been drawn up are no longer discernible, 
and they were probably nothing but little mounds of earth-: 
at any rate it was senseless to draw up a long line against the 
immense mass of enemies. The Gauls, on the other hand, 
were enabled without any difficulty to turn off to the left. 
They proceeded to a higher part of the river, where it was 
more easily fordable, and with great prudence threw them- 
selves with all their force upon the right wing, consisting of 
the civic legions. The latter at first resisted, but not long; 
and when they fled, the whole remaining line, which luitil 
then seems to have been useless and inactive, was seized with 
a panic. Terror preceded the Gauls as they laid waste every- 
thing on their way^, and this paralysed the courage of the 
Romans, instead of rousing them to a desperate resistance. 
The Romans therefore were defeated on the Alia in the most 
inglorious manner. The Gauls had taken them in their rear, 
and cut off their return to Rome. A portion fled towards the 
Tiber, where some effected a retreat across the river, and 
others were drowned ; another part escaped into a forest. The 
loss of life must have been prodigious, and it is inconceivable 
how Livy could have attached so much importance to the 
mere disgrace. If the Roman army had not been almost 
annihilated, it would not have been necessary to give up the 
defence of the city, as was done, for the city wag left unde- 
fended and deserted by all. Many fled to Veil instead of 
returning to Rome: only a few, who had escaped along the 
high road, entered the city by the Colline gate. Rome was 
exhausted, her power shattered, her legions defenceless, and 
her warlike allies had partly been beaten in the same battle, 

" It is very difficult to recognise the places in Lombai'dy where the battles of 
1799 were fought, because the roads have since been laid different^. The same 
is the case at Liitzen, Breitciifeld, and Leuthen; nay, even at Prague and Colliu 
it is not an easy matter to identify the fields of battle. — N. 

' The Gauls destroyed all the towns in Gallia Cispadana, and they themselves 
lived only in villages. When subsequently the Romans conquered the country 
of the Insubrians, they found no trace i>f the ancient population. — N. 


and were partly awaiting the fearful enemy in their own coun- 
tries. At Rome it was believed that the whole army was 
destroyed, for nothing was known of those who had reached 
Veii. In the city itself there were only old men, women, and 
children, so that there was no possibility of defending it. It 
is, however, inconceivable that the gates should have been 
left open, and that the Gauls, from fear of a stratagem, should 
have encamped for several days outside the gates. A more 
probable account is, that the gates were shut and barricaded. 
We may form a vivid conception of the condition of Rome 
after this battle, by comparing it with that of Moscow before 
the conflagration : the people were convinced that a long- 
defence was impossible, since there was probably a want of 
provisions. Livy gives a false notion of the evacuation of the 
city, as if the defenceless citizens had remained immovable in 
their consternation, and only a few had been received into the 
Capitol. The determination, in fact, was to defend the Capi- 
tol, and the tribune Sulpicius had taken refuge there, with 
about 1000 men. There was on the Capitol an ancient well 
which still exists, and without which the garrison would soon 
have perished. This well remained unknown to all antiqua- 
ries, till I discovered it by means of information gathered 
from the people who live there. Its depth in the rock 
descends to the level of the Tiber, but the water is now not 
fit to drink. The Capitol was a rock which had been hewn 
steep, and thereby made inaccessible, but a clivus, closed by 
gates both below and above, led up from the Forum and the 
Via Sacra. The rock, indeed, was not so steep as in later 
times, as is clear from the account of the attempt to storm it; 
but the Capitol was nevertheless very strong. Whether some 
few remained in the city, as at Moscow, who in their stupe- 
faction did not consider what kind of enemy they had before 
them, cannot be decided. The narrative is very beautiful, 
and reminds us of the taking of the Acropolis of Athens by 
the Persians, where, likewise, the old men allowed themselves 
to be cut down by the Persians. Notwithstanding the impro- 
babihty of the matter, I am inclined to believe that a number 
of aged patricians — their number may not be exactly histo- 
rical — sat down in the Forum, in their ofBcial robes, on their 
curule chairs, and that the chief pontiff devoted them to 
death. Such devotions are a well-known Roman custom. It 


is certainly not improbable that the Gauls were amazed when 
they fomid the city deserted, and only these old men sitting so 
immoveable, that they took them for statues or supernatural 
visions, and did nothing to them, until one of them struck a 
Gaul who touched him, whereupon all were slaughtered. To 
commit suicide was repugnant to the customs of the Romans, 
who were guided in many things by feelings more correct and 
more resembling our owu, than many other ancient nations. 
The old men, indeed, had given up the hope of their country 
being saved; but the Capitol might be maintained, and the 
survivors preferred dying in the attempt of self-defence, to 
taking refuge at Veii, where after all they could not have 
maintained themselves in the end. The sacred treasures were 
removed to Caere, and the hope of the Romans now was, that 
the barbarians would be tired of the long siege. Provisions 
for a time had been conveyed to the Capitol^ where a couple 
of thousand men may have been assembled, and where all 
buildings, temples, as well as public and private houses, were 
used as habitations. The Gauls made fearful havoc at Rome, 
even more fearful than the Spaniards and Germans did in the 
year 1527. Soldiers plunder, and when they find no hmuan 
beings, they engage in the work of destruction ; and fires 
break out, as at Moscow, without the existence of any inten- 
tion to cause a conflagration. The whole city was changed 
into a heap of ashes, with the exception of a few houses on 
the Palatine, which were occupied by the leaders of the Gauls. 
It is astonishing to find, nevertheless, that a few monuments 
of the preceding period, such as statues, situated at some 
distance from the Capitol, are mentioned as having been pre- 
served ; but we must remember that travertino is tolerably fire 
proof That Rome was burnt down is certain; and when it 
was rebuilt, not even the ancient streets were restored. 

The Gauls were now encamped in the city. At first they 
attempted to storm the clivus, but were repelled with great 
loss, which is surprising, since we know that at an earlier 
time the Romans succeeded in storming it against Appius 
Herdonius. Afterwards they discovered the footsteps of a 
messenger who had been sent from Veii, in order that the 
state might be taken care of in due form ; for the Romans in 
the Capitol were patricians, and represented the curies and the 
government, whereas those assembled at Veii represented the 


tribes, but had no leaders. The latter had resolved to recall 
Camillus, and raise him to the dictatorship. For this reason 
Pontius Cominius had been sent to Rome to obtain the sanc- 
tion of the senate and the curies. This was quite in the spirit 
of the ancient times. If the curies had interdicted him aqua 
et igni, they alone could recall him, if they previously obtained 
a resolution of the senate authorising them to do so ; but if he 
had gone into voluntary exile, and had given up his Roman 
franchise by becoming a citizen of Ardea before a sentence 
had been passed upon him by the centuries, it was again in 
the power of the curies alone, he being a patrician, to recall 
him as a citizen; and otherwise he could not have become 
dictator, nor could he have regarded himself as such. 


It was the time of the dog-days when the Gauls came to Rome, 
and as the summer at Rome is always pestilential, especially 
during the two months and a half before the first of September, 
the unavoidable consequence must have been, as Livy relates, 
that the barbarians, bivouacking on the ruins of the city in the 
open air, were attacked by disease and carried off, like the 
army of Frederick Barbarossa when encamped before the castle 
of St. Angelo. The whole army of the Gauls, however, was 
not in the city, but only as many as were necessary to blockade 
the garrison of the Capitol; the rest were scattered far and 
wide over the face of the country, and were ravaging all the 
unprotected places and isolated farms in Latium; many an 
ancient town, which is no longer mentioned after this time, 
may have been destroyed by the Gauls. Xoue but fortified 
places like Ostia, which could obtain supplies by sea, made a 
successful resistance, for the Gauls were unacquainted with the 
art of besieging. The Ardeatans, whose territory was likewise 
invaded by the Gauls, opposed them, under the command of 
Camillus*; the Etruscans would seem to have endeavoured to 

' A difficult passage in Ovid's Metamorphoses must perhaps be referred to 
this war: he says, that out of the nii)is of the town of Ardea, which had been 


avail themselves of the opportunity of recovering Veil, for we 
are told that the Romans at Veii, commanded by Caedicius 
o-ained a battle against them, and that, encouraged by this suc- 
cess, they began to entertain a hope of regaining Rome, since 
by this victory they got possession of arms. 

A Roman of the name of Fabius Dorso, is said to have 
offered up in broad daylight, a gentilician sacrifice on the 
Quirinal; and the astonished Gauls are said to have done him 
no harm — a tradition which is not improbable. 

The provisions in the Capitol were exhausted, but the Gauls 
themselves being seized with epidemic diseases became tired of 
their conquests, and were not inclined to settle in a country so 
far away from their own home. They once more attempted to 
take the Capitol by storm, ha-\ang observed that the messenger 
from Veii had ascended the rock, and come down again near 
the Porta Carmentalls, below Araceli. The ancient rock is 
now covered with rubbish, and no longer discernible. The 
besieged did not think of a storm on that side ; it may be, that 
formerly there had in that part been a wall, which had become 
decayed; and in southern countries an abundant vegetation 
always springs up between the stones^, and if this had actually 
been neglected it cannot have been very difficult to climb up. 
The Gauls had already gained a firm footing, as there was no 
wall at the top. The rock which they stormed was not the 
Tarpeian, but the arx — when Manlius who lived there was 
roused by the screaming of the geese : he came to the spot and 
thrust down those who were climbing up. This rendered the 
Gauls still more inclined to commence negotiations ; they were 
moreover called back by an inroad of some Alpine tribes into 
Lombardy, where they had left their wives and children : they 
oftered to depart if the Romans would pay them a ransom of a 
thousand poiinds of gold, to be taken no doubt from the Capi- 
toline treasury. Considering the value of money at that time. 

laid Avaste by the barbarians, there arose a heron. INIodern commentators have 
incorrectly referred this destruction to theHannibalian-ivar; it might be an allu- 
sion to some Samnito war, in which Ardea was dcstro}-ed, as we mav jtcrhaps 
infer from Strabo, who says that the Samnites carried their conquests as far as 
Ardea; but the Samnites were surely not called b;u-barians: wc jirobahh- have 
here the reverse of the tradition given in the text, that the Ardcatans uniier 
Camillus conquered the Gauls. — N. 

' Virgil says: Galli per dumus ddenint, and Livy too speaks ofvirquUa. N. 


tlie sum was enormous: in the time of Theodosius indeed, there 
were people at Rome who possessed several hundred weights of 
gold, nay, one is said to have had an annual revenue of two 
hundred weights. There can be no doubt that the Gauls 
received the sum they demanded, and quitted Rome; that in 
weighing it, they scornfully imposed upon the Romans is very 
possible, and the vae victis too may be true : we ourselves have 
seen similar things before the year 1813. But there can be no 
truth in the story told by Livy, that while they were disputing 
Camillus appeared with an army and stopped the proceedings, 
because the military tribunes had had no right to conclude the 
treaty. He is there said to have driven the Gauls from the 
city, and afterwards in a two-fold battle to have so completely 
defeated them that not even a messenger escaped. Beaufort, 
inspired by Gallic patriotism, has most excellently shown what 
a complete fable this story is. To attempt to disguise the mis- 
fortunes of our forefathers by substituting fables in their place 
is mere childishness. This charge does not affect Livy indeed, 
for he copied only what others had written before him ; but he 
did not allow his own conviction to appear as he generally does, 
for he treats the whole of the early history with a sort of irony, 
half believing, half disbelieving it. 

According to another account in Diodorus, the Gauls be- 
sieged a town allied with Rome (its name seems to be miswritten 
but is probably intended for Vulsinii), and the Romans relieved 
it and took back from the Gauls the gold which they had paid 
them ; but this siege of Vulsinii is quite unknown to Livy. A 
third account in Strabo and also mentioned by Diodorus does 
not allow this honour to the Romans, but states that the Cae- 
rites pursued the Gauls, attacked them in the covmtry of the 
Sabines and completely annihilated them. In like manner the 
Greeks endeavoured to disguise the fact, that the Gauls took 
the money from the Delphic treasury and that in a quite histo- 
rical period (Olymp. 120). The true explanation is undoubt- 
edly the one found in Polybius, that the Gauls were induced 
to quit Rome by an insurrection of the Alpine tribes, after it 
had experienced the extremity of humiliation. Whatever the 
enemy had taken as booty was consumed, they had not made 
any conquests but only indulged in plunder and devastation ; 
they had been staying at Rome for seven or eight months, and 
could have gained nothing further than the Capitol and the 


very money which they received without taking that fortress. 
The account of Polybius throws light upon many discrepant 
statements, and all of them, not even excepting Livy's fairy- 
tale-like embellishment, may be explained by means of it. The 
Romans attempted to prove that the Gauls had actually been 
defeated, by relating that the gold afterwards taken from the 
Gauls and buried in the Capitol, was double the sum paid to 
them as a ransom; but it is much more probable that the Ro- 
mans paid their ransom out of the treasury of the temple of the 
Capitoline Jupiter and of other temples, and that afterwards, 
double this sum was made up by a tax, which agrees with a 
statement in the history of Manlius, that a tax was imposed for 
the purpose of raising the Gallic ransom : surely this could not 
have been done at the time of the siege, when the Romans 
were scattered in all parts of the country, but must have taken 
place afterwards for the purpose of restoring the money that 
had been taken. Now if at a later time there actually existed 
in the Capitol such a quantity of gold, it is clear that it was 
believed to be a proof that the Gauls had not kept the gold 
which was paid to them. 

Even as late as the time of Cicero and Caesar, the spot was 
shewn at Rome in the Carinae, where the Gauls had heaped up 
and burned their dead ; it was called busta Gallica, which was 
corrupted in the middle ages into Portogallo, whence the 
church which was built there was in reality called S. Andreas 
in bustis Gallicis, or according to the later latinity in busta 
Gallico, — busta Gallica not being declined. 

The Gauls departed with their gold which the Romans had 
been compelled to pay, on accoimt of the famine that prevailed 
in the Capitol which was so great that they pulled the leather 
from their shields and cooked it, just as was done during the 
siege of Jerusalem. The Gauls were certainly not destroyed. 
Justin has preserved the remarkable statement that the same 
Gauls who sacked Rome went to Apulia, and there offered 
for money their assistance to the elder Dionysius of Syracuse. 
From this important statement it is at any rate clear, that they 
traversed all Italy, and then probably returned along the shore 
of the Adriatic : their devastations extended over many parts 
of Italy, and there is no doubt that the Aequians received their 
death-blow at tliat time, for henceforth we hear no more of the 
hostilities of the Aequians against Rome. Praeueste. on the 


other hand, which must formerly have been subject to the 
Aequians, now appears as an independent town. The Aequians, 
who inhabited small and easily destructible towns, must have 
been annihilated during the progress of the Gauls. 

There is nothing so strange in the history of Livy as his 
view of the consequences of the Gallic calamity ; he must have 
conceived it as a transitory storm by which Rome was humbled 
but not broken. The army according to him was only scattered, 
and the Romans appear afterwards just as they had been before, 
as if the preceding period had only been an evil dream, and as 
if there had been nothing to do but to rebuild the city. But 
assuredly the devastation must have been tremendous through- 
out the Roman territory: for eight months the barbarians had 
been ravaging the country, every trace of cultivation, every 
farmer's house, all the temples and public buildings were des- 
troyed; the walls of the city had been purposely pulled down, 
a large number of its inhabitants were led into slavery, the 
rest were living in great misery at Veil; and what they had 
saved scarcely sufficed to buy their bread. In this condition 
they returned to Rome. Camillus as dictator is called a second 
Romulus, and to him is due the glory of not having despaired 
in those distressing circumstances. After the time of the 
Volscian war, Rome had no longer been able to concede to her 
former allies, who were then in a state of weakness, the same 
rights as before : they had been subjects of Rome for nearly 
seventy years, though Rome was very mild in the use of her 
power. But all those people who had suffered less than Rome, 
now renounced her supremacy, and this is the defectio Latino- 
rum qui pa' centum fere annos nunquam ambigua fide in amicitia 
populi Romani fuerant, of which Livy speaks : nothing is more 
natural than that they should assert their independence. It 
would be very lamentable if unnatural regulations had an in- 
vincible power, rendering it impossible for that which is in 
accordance with nature finally to become established. It is 
quite a different question how it necessarily came to j^ass that 
shortly before the Gallic invasion the Romans in reality had 
the supremacy; this certainly was the case, as under similar 
circumstances among the seven Dutch provinces, although all 
had perfectly equal rights, yet Holland in fact stood at their 
head, and occupied the rank which belonged to it in virtue of 
Its wealth and popiilation. In like manner, Rome might be 


regarded as tlie head of the confederacy, but only so long as 
she was in possession of all her power. 

There is an ancient tradition that during the famine, the 
aged were killed in order to save them from the pangs of death 
by starvation, and to preserve the little means which yet re- 
mained for those who were to perpetuate the republic. Things 
were almost as bad as at the destruction of Magdeburg, where 
the nmnber of inhabitants was reduced from 30,000 to 3,000. 
Even after it was rebuilt, Rome must for several generations 
have been only a shadow of what it had been previously to its 
destruction. It is quite natural that the people should have 
been desponding, and that the tribunes should have insisted 
upon abandoning Rome and settling at Yeii. It is the merit of 
Camillus that he resisted this pusillanimous despondency, and 
he was on that occasion supported by his high aristocratic senti- 
ments. It required great acuteness to hit upon the right plan : 
the gods had abandoned Veii, and Juno had loudly declared that 
she would not inhabit it, but Rome. The discussions upon this 
subject in Livyhave a peculiar charm. I do not mean to say, that 
Rome would not have been able to strike new roots at Yeii, but it 
is more probable that it would have entirely perished; the 
Latins would have made themselves masters of the left bank of 
the Tiber, and perhaps a Volscian or Latin colony would have 
been established on the seven hills. The situation of Rome on a 
river between three nations had been chosen by Providence 
for her greatness ; its advantages are obvious : but at Veii the 
Romans would perhaps have become Etruscans. The senate 
now acted like a severe flither : after it had passed the resolu- 
tion to rebuild Rome, which was very hard for the poor, an 
order was issued that, for the purpose of restoring Rome, Yeii 
shoidd be destroyed. The senate, it is said, gave gratis, tiles, 
stones and other building materials, all of which were to be 
found at Yeii, the buildings of which were the property of 
the Roman republic. The materials had now to be carried to 
Rome. The new habitations were badly built huts, and it was 
only gradually tliat better houses were erected. The senate 
gave the people leave to build as they pleased, for according to 
Roman principles all private property had during the confusion 
reverted to the state, which now gave permission to occupy it 
anew. The walls were restored, and the dangerous place in 
the Capitol where the Gauls climbed up, was strengthened by 

VOL.1. T 


a substructure of square blocks. It was not till the time of 
Augustus that Veii was restored as a military colony, but it 
was only a small place like Gabii, Lavici and others. 

The longer I have been engaged in these investigations, the 
more satisfaction have I derived from them. I am conscious 
of having searched after truth without allowing myself to be 
dazzled by authority. ^\^ien I find that statements which I 
had absolutely rejected, are after all, correct in a certain sense, 
and that they have become imperfect only through want of 
knowledge or through having fallen into oblivion, I am always 
greatly rejoiced. This has happened to me frequently, and 
especially in regard to the liistory of the Roman rate of interest 
and the laws about usury. If I am to state what I think, I 
must say that before my time these subjects were in the 
greatest confusion. During the eighteenth century, the 
antiquities of the Eoman law, especially the jus publicum, 
were sadly neglected: I except Schultiug. Heineccius, a man 
deserving of all honour, possessed great talent and learning, 
but did not know what course to take ; he laboured under the 
same mistake as the men of the sixteenth century whose 
disciple he was, and had no independence of judgment. A 
variety of opinions have been published on the Roman rate of 
interest: among others Hugo of Gottingen has written upon 
the subject: he came forward as the founder of a new school 
of learned jurisprudence; he was a man of excellent taste, 
and took great interest in these questions, but did not possess 
the solid learning which is required for such discussions. 
Savigny and I were long ago convinced, that what Hugo had 
written on the rate of interest was worth nothing, and that 
the whole subject must be investigated anew. Savigny did 
not undertake the task, but I was led to it in the course of 
my investigations into Roman history: my results have been 
confirmed by Schrader of Tubingen, and my opinions are now 
generally adopted.' Roman contracts of loan were concluded 
for years of ten months each, and one ounce was paid as 
interest upon one As, that is, one twelfth part of the capital, 
which is as much as ten per cent, in a year of twelve months. 

' As an artist opens his pupil's eye and trains it best 1a- working in liis pre- 
gcncc, so it is in science also: he who has searched all his life certainly does a 
service to his disciples if lie sho^vi; to them the manner in which he made pro- 
fjress, and sometimes also how he was obliged to retrace his steps. — N. 


Hugo thought that one twelfth was paid for every mon th 
which proves that he had no perception of what is possible in 
the affairs of practical life. Jurisprudence, in general, has 
two sides : the one is science or theory, and the other the prac- 
tice of ordinary life; in regard to the latter, we Germans are 
in a wrong way: in other countries things are better, inas- 
much as the knowledge of theory goes more hand in hand 
with the relations of practical life. It is quite remarkable 
that there are teachers of law, who have no knowledge of 
actual affairs, which appears to me as absurd as if a man v/erc 
to come forward to teach medicine without having any notion 
of disease. A practical knowledge must support historical 
jurisprudence, and if any one has got that, he can easily master 
all scholastic speculations. The later Roman law of debt was 
taken entirely from the Greek law, and the calculation of the 
syngrapha and centesimae, such as it existed in the time of 
Cicero, arose from the condition of things established in the 
Greek cities of Athens, Rhodes and Alexandria. We read in 
Tacitus, that thefoenus unciarium was introduced by the laws 
of the twelve tables, and in Livy that it was established at the 
beginning of the fifth century. These statements have been 
considered an inexplicable contradiction, and I too formerly 
beHeved that Tacitus was mistaken ; but I am now of a diffe- 
rent opinion. We must here make a distinction : it does not 
follow from Livy's account that the foenus unciarium was not 
mentioned in the twelve tables. Down to the time of the 
Gallic invasion we hear of no complaint about usury, but after- 
wards, when every one was obliged to build, the law of usury 
was probably abolished, in order to enable every one to obtain 
money on any terms. Hence arose a dreadful state of debt; 
and forty years later the ancient laws of usiu'y were re-estab- 
lished. Livy is therefore probably correct in saying that at 
one time the taking of interest was entirely forbidden. In the 
year 1807, some friends of mine in opposition to my urgent 
remonstrances, carried a decree by which the laws of usury were 
abolished in Prussia; but the consequences were very unfortu- 
nate. Afterwards the money could not be paid, and then 
faciehant vei'suram, that is, the interest was added to the capital. 
It is unaccountable how men could be found at that time to 
advance the money; it is true people were content Avith satis- 
fying their most pressing wants, and for this reason the senate 



allowed them to build as they pleased ; but however much the 
state might do to facilitate the building, still the restoration 
must have been immensely expensive. I believe that the 
means were obtained through the clientela : the grand determi- 
nation to restore Rome, which had been made by the senate in 
the consciousness of her immortality, was very imposing and 
must have made people believe that the strength of the state 
was unexhausted ; and thus capitalists in various parts of Italy 
may have been tempted to go to a place where they might 
expect to make enormous profits : the patricians had probably 
not been able to rescue such immense capitals from the Gallic 
calamity. "WTien therefore a Syracusan or a Neapolitan came 
to Rome with ready money, he was not allowed to lend it 
himself, and accordingly became the client of a patrician who 
concluded the nexum ibr him. In this manner the condition 
of the commonalty down to the passing of the Licinian laAV 
was extremely Avretched, and it was unfair that the order 
which enjoyed so many advantages in the state, should 
also derive a usurious interest from their less fortunate fellow- 


If Rome alone had been destroyed by the Gauls, as the 
reader of Livy must believe, unless he rises to a higher point 
of view, it' would be inconceivable how Rome could have 
maintained herself against the neighboiu-ing nations, which 
had seized the opportunity of shaking off her yoke. But her 
neighbours like herself had undoubtedly deeply felt the calamity, 
even supposing that they defended their towns, and that many 
may have warded off devastation by a heavy war-tax. The 
condition of a great part of Italy reminds us of the time which 
followed immediately after the termination of the Thirty Years' 
AVar,when new wars broke out forthwith. I shall give you only 
brief sketches of these events. We clearly see that the Etrus- 
cans rose against the Romans, and that this affair terminated 
favorably for the latter. Sutrium and Nepet were then the 

FOUR NEW TlilBES. 277 

Roman frontier towns towards Etruria, — all the rest, including 
Falerii, was lost, — and even these towns were sometimes be- 
sieged and actually taken; when the Romans had reconquered 
them they formed them into colonies. The war was carried 
on mainly against the Tarquinians and Vulsinians. The fact 
of the Etruscans endeavouring to recover the conquests made 
by the Romans, shews that the Etruscan confederacy also was 
then in a state of dissolution, for the northern Etruscans were 
fighting against the Gauls, while the southern ones attacked 
Rome. In the accounts of our historians, however, these 
Etruscan wars are still as full of fabulous statements as the 
former ones. At this time we ever3'where, even in Greece, see 
a dissolution of the ancient confederacies, and a tendency to 
unite into larger states. The condition of Latium was such, 
that we may say there was no longer any bond to keep the 
confederacy together. Antium, Velitrae and Circeii, whose 
colonists were either expelled or made common cause with the 
Latins and Volscians, as well as the Hernicans, were separate 
from Rome, and scarcely the nearest towns such as Tusculum 
and Lanuvium remained faithful to her. Praeneste now be- 
came an important place : the Praenestines and Tiburtines seem 
at that time to have been allied, and Praeneste may perhaps 
have been the head of a portion of the Aequians. The boun- 
dary between the Aequians and Romans ceased to be at 
Praeneste, and extended beyond it. Political relations in the 
ancient world change with extraordinary rapidity, as is most 
manifest in xVrcadia, where the three principal tribes are in the 
end completely lost sight of The union of Latium was dis- 
solved, and a portion of the Latins along with Velitrae and 
Antium rose against Rome, and so also did Praeneste with a 
portion of the Aequians. The period of Rome's supremacy 
was gone, Yeii alone was a permanent gain, and the Romans 
now admitted Etruscan places, which had already possessed 
the franchise without the suffrage, to the full privilege of citi- 
zens, and formed them into four new tribes, the niimber of 
which thus became twenty- five. Livy erroneously states, that 
the new tribes were forjncd of those who revolted to Rome in 
the earlier wars: this is impossible, for the Romans always 
formed their new tribes of a much larccr number of individuals 
than was contained in the ancient ones, since it was only in 
this manner that they could truly unite with them, and yet 


that the influence of the new tribes in the assembly might be 
limited, althoug-h individually all tribesmen were on a footing 
of perfect equality. I for my part am convinced that all these 
new tribes had formerly been sovereign towns with their terri- 
tories. The territories of Veil, Capena, Vulsinii and others, 
undoubtedly acted the part of mere spectators in the wars of 
theiY ruling towns, and surrendered to the Komans as soon as 
they appeared, without any resistance, because to whatever 
state they belonged their condition wasj equally good or bad : 
many also were neutral, as under similar circumstances we see 
was the case, in the war between Spain and the Low Countries, 
with the towns of Brabant, which paid taxes to both the belli- 
gerent powers that they might remain unattacked. By the 
destruction of a town its territory became subject to Rome, and 
it was unquestionably to these people that Rome now granted 
the full franchise, and thus recruited the reduced number of 
her citizens. The Etruscan cities undoubtedly maintained a 
very passive attitude during this change. Rome was wise 
enough to grant to her new subjects the full plebeian franchise: 
her case was like that of Jerusalem when Ezra and Nehemiah 
returned from Babylon and rebuilt the city. 

Plutarch and Macrobius mention a tradition respecting the 
reduced state of Rome, which, however, as reported by them, 
seems to be unhistorical. The city was yet without walls, 
when some of the neighbouring and very insignificant places, 
such as Fidenae and Ficulea appeared with their armies, and 
compelled the Romans to give hostages. But the hostages, 
instead of being noble virgins, were female servants ; and their 
leader, a Greek slave named Philotis, imitated the example of 
Judith, for while the troops were celebrating their unwonted 
success, and were intoxicated, she gave a signal to the Romans 
with a torch, whereupon they rushed forth and annihilated 
their enemies. This event was placed in the month of Quinc- 
tllls, consequently four months after the evacuation of the 
city; and the tradition shews at all events how much Rome 
was conceived to have been reduced. 

After tlie formation of the four new tribes Rome had again 
an extensive territory, which formed the basis of her recovery. 
At the end of this period, affairs on the left bank of the Tiber 
continued to be in the same state of dissolution as before. On 
the right bank all the territory belonged to Rome as far as 

M. MANLIUS. 279 

Sutriiim and Nepet, which were frontier fortresses^ and beyond 
which the Ciminian forest was allowed to grow wild for the 
purpose of protection. Whenever ager jmhlicus is mentioned 
at this time, it must be conceived to have been almost exclu- 
sively in those districts. The relation of isopolity probably 
existed only with the nearest Latin places, Tusculum, Lanu- 
viuni, and Aricia. I cannot here relate to you all the events 
of that period ; the detail would be entirely useless. Lectures 
like the present should only dwell upon events which are 
important in themselves and in their consequences. Livy's 
case was different since he wrote for his own countrymen. 

Of far more importance to us are the events which occurred 
in Rome itself. Avarice and usury were among the darling 
sins of the Romans; and the less they were checked, the more 
oppressive they became. Had the system of usury not been 
so excessive, the revolution which now beo-an would have 
been accompanied with less violence. A few years after the 
evacuation of the city, the distress was so great that Livy was 
ashamed to reveal it to the world, perhaps even to himself 
]\I. Manlius rose to protect the unhappy. He does not derive 
his name Capitolinus from having saved the Capitol, but 
because he lived there; for T. Manlius, probably his father, 
appears in the Fasti with this name twenty years earlier. 
The saving of the Capitol was not the only brilliant feat per- 
formed by Manlius. He was acknowledged to be one of the 
most illustrious military heroes ; and the fact of his name not 
being mentioned in the Fasti throws light upon his position. 
He is universally said to have had consilia regni affedandi, 
but Livy states that the annals contained no evidence to sup- 
port this charge, except that meetings were held in his house, 
and that benefits had been conferred by him upon the plebcs. 
It may be that he was indignant at the ruling party, because 
he had not been rewarded for his service; but it is also 
possible that his great soul was stirred up by ungovernable 
ambition, and that he indulged in the hope of rewarding him- 
self with a crown. All his actions were of a kind which the 
purest and most benevolent mind might have suggested with- 
out being under the influence of ambition. Citizens were 
every day assigned to their creditors as slaves for debt. Man- 
lius paid for them what they owed, especially for old soldiers, 
and by the sacrifice of his whole property he restored them to 


their families. He is also said to have accused the patricians 
of having appropriated to themselves the money which had 
been recovered from the Gauls. This suspicion must have 
arisen from the imposition of the tax for the purpose of 
restoring the gold which had been paid to the Gauls, since 
the levying of a tax under such circumstances, though it was 
destined for the gods, was not free from harshness and fanati- 
cism. ]\Ianlius, who thus gained immense popularity, became 
the object of the greatest hatred to the riding body. Instead 
of profiting by his hint and relieving the distress, the patri- 
cians obstinately insisted on their rights, and thus arose a 
contest between beneficent ambition on the one hand, and the 
most stubborn oligarchy on the other, as in Ireland in the 
year 1822, where, when the cattle were bled, the poor fought 
for the blood in order to satisfy their hunger, and where the 
landlords nevertheless insisted on their le^'al claims. The 
natural consequence was a very general feeling, that any 
change would be better than such a government, and that 
Maulius as a usurper might be as useful as many a Greek 
tyrant. This state of things undoubtedly became very danger- 
ous. When a government is in a bad course, and vinwilling 
to retrace its stej^s, it drives men to sin, and has much to 
answer for before God and man. The Roman government 
was in this predicament, and ordered Manlius to be arrested ; 
but this led to nothing, for a general sympathy manifested 
itself for him, who until then had committed no crime. The 
plebes put on mourning, and assembled in crowds at the gate 
of his dungeon. Tlie government therefore was obliged to 
set him free. It had acted rashly, and as ]\Ianlius was now 
provoked, it thought that he was sure to take wrong steps, 
and that it would thus obtain an opportunity of crushing him. 
Manlius had a difficult part to act. Under such circumstances 
men often begin their career with the purest intentions, but 
gradually fall into frightful aberrations. I believe that Man- 
lius did not start with the idea of making himself the tyrant 
of his country; but when the men of his own order reviled 
him, and misinterpreted his pure intentions, the germs of 
his actions became poisoned, and this might lead him to 'the 
detremination to set himself up as tyrant; but no evidence of 
that supposition is to be found. The tumiilt in the mean 
time increased, and Manlius, enraged, and proud because he 


had conquered, demanded that a portion of the domain land 
should be sold, and that the produce should be employed in 
paying the debts of the poor : a fair demand, as the state was 
the proprietor of the domain land. But the oligarchs were 
bent upon reserving the possession of it for themselves, and 
rejoiced at the wretchedness of the plebeians. The distress 
rendered the dependence of the plebeians very great. So long 
as the 'praefectus iirbi had it in his power to assign a debtor to 
his creditor, every man was in danger of losing his freedom. 
It may be that dangerous thoughts gained from day to day 
greater ascendancy over the mind of IManlius, and thus at 
last the patricians ventured to accuse him. Two tribunes 
declared for the senate; and according to Zonaras, Camillus 
was appointed dictator for the occasion. Under the terrors of 
the dictatorship, Manlius was summoned before the assembly 
of the centuries, but no one dared to imprison him again. On 
giving security, he was allowed to retain the enjoyment of his 
liberty; and on the appointed day he appeared and defended 
himself, which is the strongest argument for his innocence, 
since he might have withdrawn from the city. He referred 
to his great military achievements and services as a proof of 
his sentiments; he brought forward the spoils of thirty enemies 
whom he had slain, and forty marks of honour which he had 
received in war; he appealed to the citizens whom he had 
saved, and among them even to the magister equitum ; he 
pointed to the Capitol, which could be seen from the Campus 
]\Iartius — and the centuries acquitted him. But the oligarchy 
was not satisfied with this, and the senate summoned him 
before the curies {concilium, jwpuli), who as his peers were 
to try his case in the Petclinian grove, a fact which Livy and 
all his followers have misunderstood. As the concilium populi 
is rarely mentioned, Livy thought of a tribunician accusation; 
but at tlie same time he cannot deny that the patrician duum- 
viri were his accusers. The meeting was in the Petelinian 
grove, not because the Capitol could not be seen from that 
spot, but because his enemies felt an aversion to pronouncing 
the sentence of death in the city, and yet were obliged to 
meet in a consecrated place. IManlius was condemned and 
thrown down the Tarpcian rock. This catastrophe, like the 
death of Sp. Cassius, produced for a few years a death-like 
stillness at Rome; but the patricians had nevertheless to atone 


for tKeir crime, as was always the case, although the full 
vengeance did not fall upon them, for down to the time of 
C. Gracchus, who called the murderers of his brother to 
account S the rulers who committed such a crime were not 
made personally responsible for it ; and to this forbearance 
Rome owed the preservation of her liberty. From the blood 
of Manlius arose men, whose object was not to avenge him, 
but to accomplish what he had desii'ed. Licinius and Sextius 
had perhaps (nay probably) been his friends, and his ignomi- 
nious death gave them courage to defy all dangers in accom- 
plishing their great work. Inspired by his example, they 
performed their task without shedding one drop of blood. 

It was about ten or eleven years after the destruction of the 
city, that two tribunes of the people, C. Licinius, and 
L. Sextius, placed themselves at the head of their order, with a 
firm determination to place at length the relation of the two 
orders on a just footing: it was not their intention that the 
patricians, as a distinct order, should perish, but the plebeians 
were to stand by their side with equal rights, and the state, 
according to the original idea, was to be a double state, of two 
perfectly equal communities. The military tribunes, according 
to our authorities, were again nearly always patricians, which 
is inconceivable: something must be wanting here; but the 
excerpts De Sententiis from Dion Cassius, unfortunately contain 
nothing about this period. The patricians were satisfied Avith 
the military tribuneship, and did not Avant any consuls. There 
is a foolish story explaining the motive that induced Licinius 
to come forward in this manner; but it was easy for Beaufort 
to shew that it is nothing but a fiction. j\I. Fabius Ambustus 
is said to have had two daughters, the one married to the 
patrician Su.lpicius, and the other to C. Licinius. Sulpicius 
was military tribune, and as usual returned home with the lie- 
tors; the younger sister was staying with the elder, and being 
frightened by the noise, was ridiculed by her sister, who said 
that it was natural to be sure that the noise should alarm her, 
since she was married to a man who could never attain to this 
honour. Beaufort has justly remarked, that the children of 

' Mu-abcau said at Marseilles in the year 1789, that C. Gracchus called to 
hc.avcn to avenge the Wood of his brother, and that out of that blood M.^rius 
arose; but Gracchus was an innocent and holy man, while Marius was a 
tyrant. — N. 


M. Fabiiis Ambitstus could not possibly be unacquainted with 
this mark of honour, and it is an equally unhistorical state- 
ment, that the yovmger Fabia entreated her father and hus- 
band to procure it for her also, inasmuch as the mihtary tri- 
buneship was ^open to the plebeians as much as to the patri- 
cians, and M. Fabius Ambustus himself afterwards appears 
among those who lent a helping hand in the attempt to over- 
turn the Licinian laws. The w^iole story is a miserable piece 
of scandal, invented by a party which was annoyed at the suc- 
cess of the plans of its opponents. The motives of men are 
often really despicable, but there is no reason for coming to 
such a conclusion generally, and we ought not to trace every 
thing great to mean and contemptible motives. Livy merely 
copied the tale from others, and in his haste and want of a 
vivid conception of the circmnstances, he wrote it down, not 
intending to represent it as a real history, but only as a pretty 
story : his soul was pure and noble, and although his patrician 
predilections sometimes lead him astray, he nevertheless speaks 
truly when he says in his preface, that he was impelled by an 
irresistible power to search after that which was great in the 
early times. 

But whatever may have been the occasion, the object was 
plain enough, namely to remove the existing abuses by a 
thorough reform. The reform proposed by Licinius and 
Sextius had two great ends; and to relieve the momentary 
distress was their third object. The first bill which they 
brought forward, ordained that no more military tribimes 
should be elected, but consuls, one of whom should necessarily 
be a plebeian. The patricians, notwithstanding their small 
numbers, were still predominant in the government, and for a 
long time endeavoured to prevent the passing of the bill, until 
in the end the matter was so managed that all their intrigues 
became useless: these very intrigues rendered it necessary to 
bring forward the bill in its absolute form. The tribunes 
could not have said that the worthiest should be elected from 
the two orders, for as the curies still had to sanction the 
elected magistrate, andcovdd refuse their sanction to a plebeian, 
it was necessary to fix the appointment of one plebeian as 
indispensable. The division, moreover, was important for the 
patricians themselves, for as soon as the plebeians acquired 
power in the senate, they would have elected two men from 


their own order. It was not till two hundred years later, that 
the plebeians actually acquired this preponderance, when the 
extreme diminution in the number of patricians became sensi- 
bly apparent, the patrician being to the plebeian nobility in the 
ratio of one to thirty. The second law established the prin- 
ciple that the plebeians should have a share in the possession 
of the ager puhlicus, as well as the patricians, and that, in ac- 
cordance with the Cassian law, a portion of it should be given 
to them in full ownership, to indemnify them for the past; in 
future it was to be a rule that one part of it should always be 
assigned to the patricians as their possession, and another be 
distributed among the plebeians as their real property. No 
individual was to possess more than 500 jugera; the surplus 
was to be divided among the plebes in lots of seven jugera, and 
no one was to be allowed to keep more than a certain number 
of cattle on the common pasture, during the summer in the 
mountains, and during the winter in the meadows near the 
city. The third bill contained the temporary measure regard- 
ing the debts of the plebeians : the interest which had been 
added to the capital was to be cancelled, and the rest was to be 
paid back by three annual instalments, each year being reck- 
oned at ten months, and undoubtedly without interest. This 
was indeed something like a general bankruptcy; but the 
matter could not be settled otherwise, and the creditors had 
assuredly made sufficiently large profits by their former usut}'. 
The tribunes in this case did for individuals, what Sully, after 
the unfortunate times of the League did for the state, in di- 
minishing the amount of debt: he cancelled the usurious 
interest already paid, and allowed the remaining capital to 
stand at the ordinary rate of interest. It was in consecpience 
of this violent measure that France reached its high state of 
prosperity under Louis XIII, while previously the farmers of 
finance and the usurers had alone fattened upon the marrow of 
the nation. There is no doubt that at Rome too, it was only 
the worst individuals that suffered by the law: a gentler 
remedy would have been desirable, but none was to be found, 
and without some remedy the state would have perished. 

The patricians not only opposed these rogations with a fixed 
determination not to yield, but they also exerted all their in- 
iluence at the elections, in order that the tribunes, Avho for ten 
years were re-elected year after year, might have opponents in 


tlieir own college. The whole history of these occurrences is 
buried in utter darkness, and we cannot say from what quarter 
the opposition came, nor wherein the difficidty lay. Whether 
the tribunes themselves formed the opposition, or whether the 
patricians contrived to produce indifference and hostility among 
the commonalty, or whether the laws were passed as rogations 
by the centuries, so that only the senate and the curies refused 
their sanction, — all these are questions which we cannot answer, 
but the state of things was probably different at different 


Our authorities state that the tribunes Licinius and Scxtius, 
for the purpose of carrying their laws, opposed the election of 
new magistrates with such perseverance, that for five, or ac- 
cording to others for six years, no curule magistrates were 
elected. This is one of those accounts which we may often 
read, without being able at first to believe that they can be 
inventions; in all the Fasti we find five years, during which 
neither consuls nor military tribimes are mentioned, but only 
the tribunes of the plcbes, Licinius and Sextius; their colleagues, 
Avho surely should have been recorded along with them, are 
not named. Such also was the case in Junius Gracchanus, 
from whom the statement was adopted by Joannes Lydus, but 
it is nevertheless false. There is no doubt that the tribunes 
for a time stopped the election of curule magistrates, whereby 
the Fasti were thrown into disorder; but what would have 
been the confusion, if this had happened for five successive 
years! Interrcges were indeed sufficient for times of peace, 
but they could not have led an army into the field : and would 
the neighbours of Rome have left her undisturbed during such 
a state of internal dissolution? The story appears to have arisen 
in the first place from the certain knowledge that during the 
whole struggle the tribunes actually opposed the elections, and 
pelded only at times of the most urgent necessity, when a war 
absolutely demanded the appointment of curule magistratos; 


the periods tlierefore during which there were no magistrates 
were always short, the elections being only put off. In the 
second place, the ancients imagined that Rome was taken by 
the Gauls, in the archonship of Pyrgion^, Olymp. 98. 1, as 
they read in Timaeus, whose statement they regarded as au- 
thentic, not considering that his knowledge of the fact was not 
as certain as his statement was positive. Fabius wrote his work 
fifty Olympiads later, Olymp. 148. 1=a.u. 565 according to 
Cato ; he knew very well how people then reckoned in Greece, 
and he also knew that two hundred years previously Rome had 
been taken by the Gauls : he accordingly calculated backwards, 
but the Fasti did not agree, six or seven years being wanting 
between the taking of Rome and the Licinian rogations ; some 
time might be occupied by the interreges, who had supplied 
the place of consuls, but all the years could not be filled up in 
this manner. After the Gallic calamity, the consuls were 
elected on the kalends of Quinctilis, and in his time, perhaps 
on the kalends or ides of Augustus, for the elections always took 
place on one of these two days of a month ; by this means the 
calculation of the years changed its starting point. The result 
is, that what is senseless, is also untrue, and the Gallic conquest 
must be placed considerably, at least four years, later than the 
date usually assigned to it. Now, the first authors who pro- 
mulgated our account, certainly did not mean to say that, 
during five years, the tribunes were the only magistracy : they 
combined the Greek date and the Roman statements, but did 
not know how to find their way in the Fasti, — hence, in the 
Fasti of Varro, dictatorships are inserted, which are said to 
have lasted for a whole year, but they likewise are wrong; 
they arose merely from the shifting of the consular years; — 
the ancient authors then went beyond the restoration of the 
consulship in A.U. 388, fixed there the impossible anarchy of 
five or six years, and inserted the tribunes of the people, to 
whom however, instead of ten years, they assigned far too 
many. The interpolator found in the Fasti the title t7-ibuni,'with.- 
out any further attribute to indicate the curule magistrates, and 
therefrom he inferred the opposition to the elections, which 
Livy has spun out so much. 

' 2u/U(/)wcerToi ffxeSbj' vno nivruv, says Dionysiiis; this ffXfSb proves that all 
were not agreed, and I believe that the excellent Cincius assigned it to n diffe- 
rent year, perhaps to Olymp. 99. 1 or 2. — N. 


There can be no doubt that during these contests the Roman 
magistrates were always miUtary tribunes, and ahnost invari- 
ably patricians, on one occasion only half their number con- 
sisted of plebeians, and the presidents at the elections generally 
refused to accept any votes for plebeian candidates. The ex- 
asperation of the people rose from day to day, and went so far 
that in the end the outbreak of a civil war was feared. Under 
the dictatorship of Manlius the tribunes carried a law, which 
they had perhaps proposed along with others, that half the 
decemvirs, who were entrusted with the keeping of the Sibyl- 
line books, should be elected from the plebeians, in order to 
prevent false assertions on the part of the patricians respecting 
the prodigies. Another great advantage was gained by the 
dictator P. JManlius raising a cousin of the tribune Licinius to 
the office of magister equitum : this was certainly in accordance 
with the ancient custom, for the plebeians too had iheix equites, 
and Brutus in his time had been tribunus celerum. Wlien none 
of the tribunes made any further opposition, and the tribes had 
passed the rogation of Licinius, matters came to extremities, 
because the senate, consisting almost entirely of patricians, re- 
fused to give its sanction. The commonalty shewed much less 
obstinacy in endeavouring to carry the law respecting the con- 
sulship, which was of the highest importance to the plebeian 
nobility, than in passing the other laws. The senate here 
again tried its old tactics, attempting to get out of its difficul- 
ties by temporary concessions. But Dion Cassius relates that 
the tribunes of the people, in order to carry all their laws 
at once, combined them in one bill, and Licinius is reported to 
have said, that if they would not eat, neither should they 

In all free states there are families in which certain political 
views and principles are hereditary, for there a man is born in 
a political party as he is born in a particular church. Roman 
history furnishes many examples of this kind: the first tribune 
of the plebes was a Licinius; a Licinius was the first who led 
the people in their insurrection on the Sacred Mount ; and 420 
years later it was again a Licinius, who after the death of Sulla 
vindicated the rights of the tribuneship, so that the Licinii 
always remained the foremost among the plebeian families. 
The same observation may be made in regard to tlie Publi- 
lii and Sicinii. It may at first seem a strange limitation of 


individual freedom to be thus dependent on the principles of 
one's forefathers, as if it were an external obligation, but a little 
experience shews that it is the foundation of the firmness and 
strength of a nation. But to return to our narrative : Licinius 
then combines his various laws that all might stand or fall at 
once. Nothing is more glorious in Roman history than that 
the commonalty though far superior to their opponents in 
strength and numbers bore their machinations with the greatest 
composure and patience and without committing any illegal 
act, although the struggle lasted for a series of years. 

The aged Camillus — he was now eighty years old — was ap- 
pointed dictator : his blood had not yet been cooled, the ancient 
party-spirit and animosity still survived in him, and when 
called upon by his order he fancied he could do what was in 
reality impossible. The plebeians did not dare to resist the 
dictator, but with extreme wisdom resolved that if Camillus as 
dictator should undertake anything unlawful against them, 
they would accuse him after the expiration of his office and 
propose that he should be fined 500,000 asses. This declara- 
tion paralysed Camillus, and the senate was afraid to let matters 
come to extremities. Camillus found that he could do no more 
than Cincinnatus ninety years before,who had to avenge a dis- 
graced son. The patricians began to reflect, and Camillus 
himself advising them to yield, made a vow that he would 
build a temple of Concord if he should succeed in reconcilinrr 
the two orders. This temple was consecrated though not till 
after the death of the great man. The Romans of a later time 
thought its ancient magnificence too mean; in the reign of 
Augustus its place was supplied by another, and Trajan built 
a still more magnificent one instead of the second. Down to 
the year 1817 its site was sought for in a wrong place: it stood 
in a corner below the Salita which leads from the arch of Sep- 
timius Severus to the Capitol; several votive tablets were 
found there behind the church of S. Servius, which Pope Cle- 
ment VII. erected on the site of a more ancient church. The 
pillars of the later temple were of Phrygian marble, wrought 
Avith extraordinary elegance. Trajan loved to dwell in past 
ages: he coined Roman denarii, bearing on one side his own 
head and on the reverse the emblems of great families which 
had become extinct (for in the earlier times the rijiht of coininff 
was not an exclusive privilege of the state) : and there still 


exist a considerable number of these nummi rcstituti. It was 
the same feeling which prompted him to restore the ancient 
temple of Concord, for the spot on which the golden age of 
Rome had begun was sacred to him as it was to his friends 
Pliny and Tacitus. Its site is now clear of rubbish and is a 
classical spot in Roman topography, — the symbol of a free and 
equal constitution. 

The reconciliation was brought about in this manner: it was 
agreed that one of the consuls shovdd be a plebeian and the 
other a patrician; the ancient consulship, however, such as it 
had existed previously to the decemvirate, was not to be I'e- 
stored, but the praefect of the city was to be a permanent and 
new curule magistrate under the name o^pj'aetor wbanus.^ This 
praefectura urbis had existed even before the decemviral con- 
stitution, and was to have received a different character in that 
constitution; there were now many reasons for the patricians 
not allowing it to fall into the hands of the plebeians, because 
the whole possession of the agei^ publicus was dependent upon 
it. If for example a father bequeathed 400 jugcra to his son, 
who already possessed 400 j ugera, a conscientious praetor might 
take from him the 300 above the legal quantity; but if a 
praetor was determined to render the law inelFectual, he 
assigned the bequest to the son and would not listen to the 
charge that he already possessed more than the law allowed. 
In addition to this, it must be remembered that the laws were 
still under the superintendence of the pontiifs, and that accord- 
ingly the patricians, who alone were eligible to the pontificate, 
might say that they were exclusively entitled to be invested 
with the praetorship. Another no less important right of the 
praetor was that of appointing the judices. The centum viri, 
who were elected by the tribes, had to decide in questions of 
property, bvit all criminal cases were brought before the praetor. 
When the crime was a delictum maiiifestum, the perpetrator 
was dragged before his tribunal obtorto culfo, and the praetor 
at once determined the punishment; but when the matter was 
disputed, the praetor delegated a judex, and directed him to 
decide the case in this or that manner according to the result 
of the investigation ; there is no doubt that he himself also 
might act as judex, but he alone could not possibly have 

- This name was not devised to distinguish him from the pnicfor pere(jri»iis, 
a point in which I myself was formerly mistaken as well as many others. — N. 
A'OL. I. U 


manan-ed all the cases tKat were brought before lilm. These 
judices or judges, then and for a long time afterwards, were 
chosen from among the senators; and hence it was of great 
consequence to the patricians to reserve for themselves the 
exclusive right to the praetorship. This circumstance also 
shews the importance of the measure brought forward by Grac- 
chus. The patricians retained the possession of the praetorship 
for thirty-two years; but when a great portion of the ager 
puhlicus had passed into the hands of the plebeians, when con- 
sequently the praetor changed his character, commanded armies, 
and often perfonned the functions of a consul, the office could 
no longer be withheld from the plebeians. It should, more- 
over, be observed that the praetor was called the colleague of 
the consuls and that he had six lictors, as the two consuls 
together had twelve. 

It is flirther mentioned that the curule aediles were then 
for the first time appointed for the purpose of conducting the 
public games ; the plebeian aediles are said to have refused to 
give expensive games for the celebration of the peace, and as 
some patrician youths generously undertook to do so, the new 
office is stated to have been instituted to honour them. Even 
in the first edition of my Roman history I shewed the folly of 
this opinion; the curule aediles were neither more nor less than 
what the ancient qiiaestores pairicidii had been : they brought 
public accusations before the popular courts in cases of poison, 
sorcery, and the like. Their jurisdiction was quite different 
from that of the praetors, and when the law had not fixed a 
particular punishment for a crime, they determined the punish- 
ment according to the nature of the offence. On this subject 
the ancients entertained different notions from ourselves. I 
know the advantages of our own times, and he whose soul is 
completely absorbed in one period is not fit for any other. A 
person who looks with fondness upon past ages and would fain 
recall them, is not a homo gravis, but is diseased in his mind. 
I would rather see a man preferring the present to the past; 
but the legislative conceit of our age is very injurious, for 
legislators imagine that they can determine everything. I was 
once present in a country when the discovery was made that 
there existed a conspiracy of men who dug up corpses from 
their graves after they had been buried for many years, and 
as the law had made no provision for such a crime the monsters 


escaped with impunity. One year after the institution of the 
curule aedileship the plebeians gained access to it also, and for 
a period of 130 years there were alternately one year two pa- 
trician, and the next, two plebeian aediles. The ludi Romani 
were increased by a fourth day for the plebeians, who had 
before had their own games. From the statements made by 
Dionysius after Fabius at the end of his seventh book, it is clear 
that until the time to which those statements refer, the state 
had annually provided a large sura of money to defray the 
expenses of those games, but that in consequence of the unfor- 
tunate events in the first Punic war, the expenses were thrown 
upon individual citizens. The games were thenceforward 
given at the expense of individuals, and the cumle aedileship 
became a liturgy in the Greek sense : the aediles obtained access 
to all the great offices, but in return they were obliged to 
defray out of their own means the expenses of the games. 

The plebeian aediles were a general Latin magistracy, as is 
evident from the fact of their being mentioned as existing in 
Latin towns; but we cannot say whether the curule aedileship 
had existed before as such a local magistracy among the pa- 
tricians, or whether it was then newly created. These cirrule 
aediles have hitherto always been considered as a sort of police 
magistracy; it is true, to some extent they were so, and in so 
far, they competed with the plebeian aediles; but their real 
office did not consist in the superintendence of the corn trade, 
public buildings and the like, in which they cannot be distin- 
guished from the plebeian aediles, but they were the ancient 
quaestures parricidii who instituted their inquisitions before the 
people, as I have proved by several examples. I suspect that 
the triumviri copitales were a detached branch of the aedilician 
power. The aediles had no lictors and no imperium. Now, 
how did it happen that these new magistrates were elected in 
the comitia tributa ? It seems probable that at first they were 
elected alternately by the comitia tributa and curiota, and that 
the comitia which did not elect had to sanction the election; 
but when the j\Iaenian law reduced the sanction of the curies 
to a mere matter of form, the election was altogether trans- 
ferred to the tribes. The inferior magistrates, such as the 
triumviri monetales, quatuorviri and others, were not instituted 
till after the Hortcnsian and Maenian laws, when the curies 
had ceased to meet, and the election was altogether transferred 

U 2 


to the tribes. As regards the praetor, there can be no doubt 
that, like the consuls, he was elected by the centuries; for it is 
said that he was elected iisdem auspiciis, and tlie auspices were 
taken only for the comitia of the centuries and curies. Thus 
the few points which are known help us in explaining what is 
mysterious in the history of the Roman constitution. 


According to Joannes Lydus (that is, according to Graccha- 
nus), the government at the close of these disturbances was for 
a time in the hands of triumvirs. I shall endeavour to explain 
this elsewhere, but the fact itself is quite credible. The cir- 
cumstance that Varro in his work inscribed to Pompey, De 
Senatu habendo, mentioned the triumviri reipuhlicae constituendae 
among those who had the right to convoke the senate, is a 
strong argument in favour of it: the later triumvirs probably 
adopted the title witli reference to this early magistracy. I 
will however not deny that the first military tribunes were 
likewise called triumviri reipuhlicae constituendae in the ancient 

When the Licinian laws were passed and the first plebeian 
consul had been elected, circumstances arose which threatened 
to throw everything back again into confusion, for the patri- 
cians refused to sanction the plebeian consul. It was only 
with great difficulty that matters were settled: the patricians 
in the end yielded and recognised L. Sextius as plebeian consul. 
In this manner the lawful and necessary revolution was brought 
to a close: it had proceeded like the normal changes in the 
human body when a yoiith passes into the age of manhood. 
We cannot wonder that the peace was not cordially meant; the 
patricians yielded only to necessity, and with the firm deter- 
mination to recover what was lost as soon as an opportunity 
should oflTer. The struggle was renewed about eleven years 
later, in A.U, 399 according to the Catonian era which is adopted 
by Livy also ' ; the patricians again succeeded in obtaining 

' Chronology is here very uncertain on account of the shifting of the time at 
v.liich the magistrates enterctl upon tlifir olliie ; it was not till the time of the 
J'linic wars tliat the consuls regularly entered upon their oflice in spring, and 
during the latter perio I of the republic on tlie first of January. — N. 


possession of both places in the consulship, and continued the 
contest until A.U. 413, usurping during more than one third of 
that period the consulship for themselves exclusively. But in 
the end they were obliged to yield with disgrace, and during the 
struggle itself tliey had to make concessions to the plebeians, 
which the latter would not have demanded with such vehe- 
mence, if the peace had been honestly observed. 

The beginning of the period which now opens is marked by 
very few events; it may be, as Livy says, that no wars were 
carried on, in order not to give the plebeian consul any oppor- 
tunity of distinguishing himself, but it is also possible that 
this is merely a conjecture of his. All tlie care of the Romans 
was directed to their internal affairs, for it is natural to suppose 
that the innumerable arranciements which the Licinian law 
rendered necessary engrossed all their attention. The whole 
of the ager publicus had to be measured and divided, a 
commission was engaged in regulating everything connected 
with the debts, and a variety of other business had to be 
settled. The general assignment of land to the plebeians 
must be regarded as the cause of the rebuilding of the city. 
We shall not easily find so speedy a recovery in history, for 
Rome appears regenerated, although almost every year is 
marked by wars: a part of the debts remained, and the law 
of nexum was not abolished, but it became less and less oppres- 
sive. The changes produced by the Licinian laws must have 
been much more extensive than we are aware, and the chest of 
the patricians now probably became the common treasury of 
the republic. The time was outwardly one of tranquillity, the 
Latins, separated from Rome, lived in peace, and none but 
isolated towns, such as Tibur and Praeneste were hostile, and 
that more from mistrust than from any other special reason. 
The Tarquinians were the only enemies that really threatened 
Rome. In the year A.u. 393, thirty years after the first inva- 
sion, however, there appeared a new enemy at a distance, the 
Senonian Gauls. Whatever is said of an earlier appearance of 
the Gauls is contradicted by Polybius, who mentions all their 
expeditions, and calls this one the first subsequent to the de- 
struction of Rome. It appears that after that event the Gauls 
marched into Apulia, and there concluded a treaty with 
Dionysius of Syracuse; after plundering the country they 
returned to their own homes, the modern Romagna and 


Urbino. But a new host came across the Alps and advanced 
as far as the Anio. We must not suppose those Gauls to have 
been very warlike when they had the means of a peaceable 
existence. On the Anio, Manlius Torquatus is said to have 
fouirht in sins-le combat with a Gaul and to have taken from 
him a golden chain: this seems to be historically established, 
and we have no reason to consider it as a fable ; a great battle 
was not fought there, and the Romans though prepared were 
now on the alert and cautious. The Gauls then fixed them- 
selves in those parts, took possession of the Alban mount 
and the hills of Latium, and sallying thence laid waste the 
Latin territory; they advanced beyond Tivoli*^ into Campania, 
nay, according to one account, even as far as Apulia; they 
must consequently have subdued the Samnites, and have 
marched through their long and narrow territory, as the 
Romans did afterwards. 

These occurrences, like the Yolscian war a hundred years 
before, were followed by consequences which were highly 
advantageous for Rome. The Romans themselves, as well as 
the Latins and Hernicans, arrived at the conviction that by 
separation they were exposing themselves to great danger. 
There was no hostility between the Romans and Latins, but 
between the Romans and Hernicans there was an open war, in 
which the Romans may have taken the strong town of 
Ferentiuum: the war ended in a restoration of the ancient 
relation. The statement that the Hernicans surrendered, is 
false, for even half a century later they continued to receive 
one third of the booty, or a compensation in money until C. 
Marcius subdued them. The Latins and Hernicans united 
with Rome, and a new state was formed, as Livy relates in 
two passages'' without rccognizhig the connection. There is 
every appearance that the Latins did not yet form a compact 
state : it was impossible for them to recover the position which 
they had formerly occupied, since a great many of their towns 
had been destroyed by the Acquians and Volscians or by the 

' In the neighbom-hood of Tivoli, I have diseovercd traces of several towns 
whicli are not generally known, and wliich may have been destroyed at that 
time. Tliey arc bnilt upon hills in the form of si^uares and exhibit no traces of 
having been surrounded by walls. They shew how small were the towns, which 
were then scattered over Italy, tlicy niny have contained about fifty houses. — N. 

' Probably vii. 12 and viii. (> and 8 ; but there are also some other passages 
in wliich this is alluded to. — Ed. 


Gauls. But the Yolscians, their former enemies, were now 
likewise broken up into several states; the Antiatans seem to 
have stood by themselves, while other Volscian towns united 
themselves with Latium ; they felt an urgent need of joining 
some other state, as they were hard pressed by the Samnites 
who were makmg conquests on the Upper Liris, had taken 
Fregellae, and remained in the possession of Casinum. Thus 
a new Latin confederacy was formed, which was joined by the 
Latin colonies and a part of the Volscians, for the Romans 
seem to have renounced all claims to supremacy over the 
Latin colonies; and Sutrium and Nepet on the left bank of 
the Tiber likewise joined the Latin league. Forty-seven 
tribes, it is said, took part in the sacrifice on the Alban mount: 
a statement which must be referred to this time when Latiiira 
stood by the side of Rome as a powerful state. Another 
portion of the Volscians was admitted to the Roman franchise, 
apparently to form a counterpoise to Latium, for two new 
tribes, situated on the Volscian frontier were formed, just as in 
the treaty of Spurius Cassius, the Latins had ceded to the 
Romans the Crustuminian territory. The year A.U. 397 is 
thus remarkable for the restoration of the ancient relation 
between Rome, Latium, and the Hernicans. Festus, in the 
article Praetor ad portam, which is derived from Cincius, speaks 
as if the Romans had always been in an equal alliance^ with 
the Latins ever since the fall of Alba. This is correct in 
regard to the periods from the peace of Sp. Cassius down to 
the year A.u. 290, and from A.u. 397 down to the consulship 
of Decius ]\lus, but the intervening period is overlooked. 
Cincius undoubtedly had the correct statement, but was pro- 
bably misunderstood by Verrius Flaccus. The different times 
must here be very carefully distinguished ; 1 myself have been 
in error for many a year in regard to this point.^ A Roman 
and a Latin impcrator had in alternate years the command of 
the imited armies, he offered the sacrifice on the Capitol at 
Rome, and was saluted at the gate of the city. 

* The triumph on the Albiin mount -which is fu-st mentioned in the case of 
Papirius Maso, after the first Piiiiic war, is commonly regarded as an iuliitrary 
act of the generals, to whom the triumph at Eome was refused: but it assuredly 
was a recollection of the ancient usage, according to which the Latin coumiander 
triumphed on tlic Alban mount, and the Eoman commander at Rome. At the 
time when there were no Latin generals, the impcrator. as general of the ;dlies 
assumed the triumi)h on the Alban Mount, when the honour was refused to him 
at Rome. — N. 


The new alliance of the three states undoubtedly arose from 
a fear of the Gauls who were very near, thovigh they did not 
appear on the Tiber that year. It would be of no advantage 
to relate here the details of the war, you may read it in Livy, 
whose work you cannot study too m\ich, both as scholars and 
as men who seek and love that which is beautiful. His faults, 
which we cannot deny, are like the faults of a bosom friend 
which we must know but towards which we ought not to be 
unjust, and which ought not to disturb our feelings. It was 
a fearful time for the Eomans; the struggle with the Gauls 
continued till A.U. 406 and 407 ; and Latium and Campania 
more particularly were for thirteen or fourteen years continu- 
ally ravaged by the barbarians. On one occasion they appeared 
at the Colline gate: the Komans successfully resisted them, or 
the fight remained at least undecided ; it was the same spot 
where afterwards Sulla defeated the Samnites, and is now 
within the city. It is a continuation of the Quirinal hill 
which slopes downwards; on the left side there is a deep 
valley, and where the Quirinal comes down to the plain, other 
hills again arise, over which run the walls of the city : it was 
undoubtedly on these latter hills that the Gauls and Samnites 
were encamped. Whoever of you has the happiness to visit 
Borne may heighten it by making himself acquainted with 
these localities. 

One of the changes which were brought about by this new 
alliance with the Latins, is expressly mentioned by Livy, and 
was, that New Latium was governed by two praetors^ whereas 
Ancient Latium had been governed by a dictator, as we know 
from Cato (in Priscian). An alliance between the liomans 
and Samnites, which is likewise mentioned by Livy, belongs 
either to this or to a somewhat later time. We may indeed 
suspect that such connections existed between the Samnites 
and Romans even at an earlier period; but we cannot assert it 
with certainty, in consequence of the vagueness of a statement 
in Fcstus in the article Numerius. According to this passage, 
one of the Fabii, who after the battle on the Cremera was sent 
as a hostage to the Gauls, married the daughter of a Samnite 
of Beneventum. Now the connubium could not have existed 
Avithout treaties. It is, however, possible, that this relation 
existed only between the Sabincs and Romans, and that the 
former transferred it to their Samnite colonics. There may 


have been two motives for forming such an alliance. If fear 
of the Gauls led to it, it must have been concluded between 
the second and the third expedition of the Gauls, that is, 
between the one to the Anio and the one to the Alban moimt; 
but according to a very probable conjecture, the alliance may 
have been the consequence of a jealousy of the power of 
Latium; for the latter country, by the addition of Volscians 
and Aequians, had become so powerful, that Rome had reason 
to be jealous. The Latins were in close contact with the 
Samnites on their frontier, and the latter were endeavouring 
to make conquests on the Upper Liris. Hence an alliance 
between Romans and Samnites was very natural: Rome and 
Latium were allied indeed, but without trusting each other. 
It is not necessary, however, to regard such a connection as a 
defensive alliance, of which, in fact, it bears no appearance 
wliatcver. It was a treaty rather than an alliance; and we 
must especially remember, that such treaties in antiquity 
usually contained an honest clause, fixing a line up to which 
each party was to be allowed to make conquests. Such was 
tlie treaty of Rome witla Carthage, that of the Carthaginians 
vinder Ilasdrubal in Spain, and that of tlie Romans witli the 
Aetolians. The moral reflections with which the division of 
the new world made by pope Alexander VI. between Spain 
and Portugal has been censured, are idle declamations ; for 
this division was nothing else than the fixing of limits to con- 
quests which each party might make. In like manner, a 
boundary was fixed in the first real peace between the Romans 
and Samnites, and the fact of the limit not being determined 
with sufficient distinctness, gave rise to the second war. 

Notwithstanding the general peace with the Latins, the 
Tiburtines acted in a hostile manner towards Rome. They 
seem to have formed an independent state, and took Gallic 
armies into their pay. A war with the Tarquinians led the 
Romans into Etruria along the sea coast. It was carried on 
with great exasperation. The Etruscans penetrated to the 
neighbourhood of Rome, but the plebeian consul, C. Marcius, 
completely defeated them, and compelled them to conclude a 
long truce. 

The internal distress continued in consequence of the mag- 
nitude uf the debts. One commission was appointed after 
another, terms were fixed, and the state liad again to interfere. 


The republic, which was now in the receipt of the tithes from 
the domain land, was so wealthy that it was in a condition to 
make some general rcg-ulation. The debts were examined by 
a commission, and all those who were involved, but could give 
security, received advances from the public treasury to pay 
their debts, a wise measure; for by paying back the capital 
the rate of interest was brought down, money accunudated 
greatly, and people were obliged to make the best use of it 
they could. On the other hand, it was determined that who- 
ever had property should not be compelled to sell it, which 
would have lowered the price of land, but that he should be 
allowed to give up his property for the debt according to a 
fair valuation. In consequence of this measure the price of 
land necessarily rose, and the rate of interest again fell: 
the financial calculation was extremely wise and subtle. It 
produced permanent and excellent results, although fresh 
misfortunes were soon followed by fresh distress. Whenever 
the calamities of a period arise from extraordinary events, 
even the wisest ruler cannot prevent the pressure and misery 
that flow from it. The misfortune to which I allude is the 
third Gallic expedition in the year a.u. 405, which was far 
more formidable to Kome than the second. The Gauls 
appeared at the gates of the city, but the Komans did not 
dare to ofier them battle. Their tactics were now greatly 
developed, yet they were wise enough to confine themselves 
to the defence of the city, although their territory was laid 
waste in consequence. The Gauls remained in Latium for a 
long time, and even during the winter. If we may believe 
the accounts of the Komans, the Gauls were in a situation 
similar to that of the Ostro-Goths under Radagasius, whom 
Stilicho confined among the Apennines^, not far from Fiesole. 
They are said to have withdrawn to the Alban hills, that is to 
Monte Cavo. It is indeed possible, but highly improbable, 
that they should of their own accord have gone to snow- 
covered hills. It is clear that L. Furius Camillus, a nephew, 
not a son of the great Camillus, marched out against the 
Gauls, and distinguished himself as a general. He was indeed 

* Even now the name given by the peasants to these mountains refers to 
tliat Gothic jjcriod. — N. (Monte Sa*:so di Cai>tro, above Miigcllo, is the moun- 
tain to the name of whicli Niebulu- liere refers, according to a conjecture of tlie 
Editor of the third vohmie of tlie Ilunian liistory, p. 79, n. 144.) 


an obstinate patrician, who violated the peace between the 
two orders; but he was nevertheless bono 'publico natus. We 
see that the Romans and Latins together sent a great army 
into the field. They formed ten legions, a number which 
could not have been furnished by the Romans alone. The 
campaign against the barbarians was conducted with great 
skill, for the Romans did not fight a battle, but thi'ew them 
into extreme distress by means of entrenchments. The state- 
ment of a grammarian that the Gauls concluded a treaty with 
the Romans must probably be referred to this time. They 
were allowed to depart, and having spread over Campania 
and plundered it, they proceeded farther south. 

Many important changes took place in the beginning of the 
fifth century. "We find it mentioned as early as the year 
A.U. 397, that the tribes declared war. This right at first 
belonged to the curies, afterwards to the centuries, and now 
to the tribes. It was natural, that as the vital power of the 
state increased, the old customs should be set aside: as, for 
example, to stop the proceedings of the assembly in conse- 
quence of lightniag, or because a bird of ill-omen fiew by, and 
the like. Such things had hitherto prevented an army being 
formed, or any resolution whatever being passed by the cen- 
turies; and it was reasonable to transfer the declaration of 
war, and other important matters, to the assembly of the 
tribes, an institution which from the beginning had been 
conceived in a purely practical sense, and adapted to the 
actual wants of the community. 


The extension of the rights of the plebeians is connected with 
the name of C. Marcius Rutilus, the first plebeian censor and 
dictator : he preserved the peace between the two estates ; and 
in his case we perceive a change in the mode of electing a dic- 
tator which is alluded to by Zonaras, but entirely overlooked 
by Livy. Up to this time the dictator had alwajs been elected 
by the patricians, that is, they elected one from among those 


candidates who were proposed, as is expressly attested by a 
passage in Livy: the last dictator elected by the curies was 
Sulpicius, for otherwise there would have been no reason to 
make particular mention of it. Livy merely copied thought- 
lessly: he has many such statements, which seem superfluous, 
unless we know from other sources how to account for them. 
Three years later, we find a plebeian dictator whom the curies 
would never have sanctioned. The change consisted in this : 
the senate only determined that a dictator should be appointed, 
and the consul named him. This is also implied in the state- 
ment of Dionysius, which he applies to an earlier period, that 
the appointment of the dictator was for a time left to the dis- 
cretion of the consul : I have sufficiently explained this subject 
in the first volume of the new edition of my Roman History. 
Thus in proportion as the curies lost power, the senate gradu- 
ally acquired an influence which it had not formerly possessed. 
The traces of the very violent commotions, which took place 
at that time, are much obscured, but a mention of them is pre- 
served in Cicero, who relates that Popilius Laenas, in his con- 
sulship, quelled a sedition of the plebes, whence he received a 
surname. I place this consulship immediately before the elec- 
tion of the plebeian dictator. In the year a.u. 400 the patri- 
cians succeeded in setting the Liciuian law at defiance and con- 
tinued to do so for a few years. Another great change took 
place, by which the appointment of a number of tribunes of 
the soldiers was assigned to the tribes. 

In regard to Etruria, it is related that in consequence of a 
truce the town of Caere was obliged to give up a portion of its 
territory; a war therefore must have taken place with Caere, 
which had never happened before ; this war is commonly much 
declaimed against as being vxngrateful on the part of Eome, 
since during the Gallic war, Caere had protected the sacred 
treasures of the Romans: but we know nothing certain about 

We have now come to the time when, as Livy says, major a 
hinc bella narranda sunt, for large masses meet each other in the 
field, and Rome has to fight with a great people which showed 
an heroic perseverance, possessed great generals and excellent 
armour (which the Romans themselves adopted from them), 
and had all the political virtues calculated to render a nation 
iUustrious in the history of the world. The struggle for life 


and death lasted for seventy years, and was interrupted only by 
treaties of peace or rather by truces. The Samnites show how 
much may be gained by a nation for its descendants by heroic 
perseverance, even when in the end it succumbs ; for the lot of 
the Samnites was always more bearable than that of many other 
nations wh.ich were subdued by Eome. Had their descendants 
limited their wishes according to their actual circumstances, 
had they not aimed, though with great heroism, at impossibili- 
ties, and not given themselves up to antiquated feelings, they 
would not have perished, no not even under Sulla. At that 
time their fate was fearful ; but only because they had ceased 
to take their own circumstances into consideration. 

The great event Avhich marks the transition of Rome from 
the age of boyhood to that of youth, was the taking of Capua 
under its protection ; but the account of this event is very ob- 
scure, and has moreover been falsified by the Romans them- 

When in antiquity we hear of a colony committing acts of 
hostility against the mother country, we always think of rebel- 
lion and ingratitude: the ancients themselves, that is, our 
authors, see in sixch an insurrection the strife of a daughter 
against her mother. In some cases indeed this view is correct, 
but in most of them, especially in the history of Italy, the 
relation is quite different. We must remember how colonies 
arose, how a portion of the territory was set apart for and 
assigned to the colonists, the remainder being left to the 
ancient inhabitants, and how the colonists then became either 
the representatives of the ruling state, or, if they emancipated 
themselves, an Independent sovereign power. The Romans 
always connected their colonies closely with themselves, and 
the same appears to have been done by the Latins. The Greek 
colonics have scarcely any resemblance to them in this respect. 
The Greeks mostly sent their colonies into desert districts, 
where they built new towns into which they afterwards some- 
times admitted pale-burghers and aliens; but they remained 
quite foreign to the nations among whom they settled, as was 
the case in Libya, on the Black Sea, in Asia j\Iinor, Thrace, 
Gaul and Spain. It was only the Pelasgian nations in Italy 
and Sicily that were akin to them, and hence the rapid growth 
of the Greek colonies in those countries. The cause of send- 
ing out a colony was usually of a political nature; it generally 


consisted of political malcontents or of tlic surplus of an over- 
populous place, and soon emancipated itself, retaining towards 
tlie mother-city only tlie duties of respect. The Roman colo- 
nies, on the other hand, were always in j^airia poiestate, and 
were bound to jDcrform certain duties. 

The system of the Samnites, and perhaps of all the Sabine 
states, was different. As they had a quite different religion, 
different fundamental forms of division, and different armour, 
so they had a different law in regard to their colonies also. 
Strabo mentions the tradition of the Samnites respecting their 
origin; they were descended from the Sabines, and found 
Oscans in the country which they conquered. That whole 
country was inhabited by Oscans, while the coast was occupied 
by the Pclasgians who at one time, we know not when, spread 
over the midland district also. At first the Pelasgians proba- 
bly dwelt from the Tiber as far as mount Garganus, but the 
Oscans, being pressed upon by the Sabines spread from the 
mountains of Abruzzo over those districts, which the Sabines, 
the ancestors of the Samnites, subsequently occupied, and 
penetrated to the southernmost parts of Italy, destroying in their 
progress the original population. Their colonization, therefore, 
was undertaken, not like that of the Romans, with a view to 
establish their dominion, but in consequence of a superabund- 
ance and fulness of life, whence we nowhere find any trace of 
a connection between the Sabine colonies and the mother- 
people. Thus it is with the Picentians, the Marsians, Marru- 
cinians, Pelignians, Vestinians, and also with the Samnites. The 
last-mentioned people consisted of four tribes which formed a 
confederacy, the Pentrians, Caudines, Hirpinians, and probably 
the Freutanians. The Frentanians were afterwards separated 
from the rest, and in their stead another canton, probably 
the Alfaterians, between Surrentum and the Silarus, was 
admitted into the confederacy. From the Samnites, again, 
other tribes issued, as the Lucanians; and out of a mixture 
of the Lucanians with Oscan and Sabellian adventurers and 
freedmen, there arose the Bruttians. When the Sabines had 
established themselves in the middle valley of the Vulturnus, 
they extended into Campania also, the most highly favoured 
country of Italy ; an Etruscan colony had existed there ever 
since the year A.u. 280. The earliest inhabitants of that 
country were imdoubtedly Tyrrhenians, whence the origin of 


Capua like that of Rome was referred to the Trojans; the 
Tyrrhenians were subdued by the Oscans, and the hitter again 
by the Etruscans: under the latter, Capua is said to have been 
called Vulturnum. The Oscan population must have been 
very numerous, for it gave a different character to the whole 
nation. But the greatness of the Etruscans lasted only a short 
time, for on the Tiber they were declining as early as the year 
A.u. 320, and consequently in Campania even much earlier. 
Now it is not surprising that Capua, a mere settlement of an 
oligarchic nation, could not maintain itself against a conquer- 
ing people, as the subdued Oscans were not very zealous in the 
defence of their masters. The Etruscans in Capua, therefore, 
made an agreement^, by which they admitted a Samnite colony, 
the epoeci of their enemies, — a foolish arrangement which we 
meet with very often in ancient history : in this manner the 
Amphipolitans admitted the Chalcidians, and the latter after- 
wards expelled the ancient Athenian colony: many similar 
examples are mentioned by Aristotle. Such towns, in which 
the ruling body of citizens consisted of different nations, 
rarely had the good fortune enjoyed by Rome, that their 
separate elements became equalized. The Samnites conspired 
against the Etruscans, and shortly afterwards, with a faithless- 
ness and cruelty peculiar to all the Sabelllan and Oscan nations, 
murdered them and kept the town for themselves. Three 
3'-ears later the Samnites spread as far as Cumae, and conquered 
that city which had long been the most illustrious place in 
Italy. The ruling population at Capua accordingly consisted 
at first of Etruscans, and afterwards of Samnites, but with a 
very numerous Oscan commonalty; for according to this 
system of colonization, a branch of the conquering nation 
received the sovereignty in the colony, one portion of the 
ancient inhabitants in the towns became clients, and the others 
remained free; whereas in the country, the population were 
made serfs as in the conquests of the Franks and Longobards. 
The relation of the Spanish colonies in Mexico likewise is of 
a similar nature; for there too the ancient population has re- 
mained. Such was the condition of Capua. We are now 
told in Roman history, that the Campanians requested succour 
from the Romans and Latins against the Samnites; but how 
could this colony have fallen out with the mother people? 
This can be explained only in the following manner. The 


commonalty, consisting of Oscans who were kept in a state of 
dependence by the Samnites, gained strength and increased: 
and while the Roman plebes gradually became united with the 
patricians, the commonalty of Capua broke out in open rebel- 
lion and crushed the Samnite patricians. This was the cause 
of the enmity between Capua and Samnium, but the Samnites 
at Capua do not appear to have been annihilated, but only to 
have lost the government : they are the Campanian cquites 
mentioned by Livy, to whom the Eoman citizens paid an 
annual tax, either as a compensation for the ager Falernus, or 
as a reward for their fidelity to Rome. The Romans were 
fond of keeping dependent people under an oligarchical 

The Samnites at that time extended from the Adriatic to 
the Lower Sea. No ancient author describes their constitution, 
and it is only from analogy and a consideration of particular 
circumstances that we can form the following probable conclu- 
sions. They consisted of four cantons, which constituted a 
confederacy, perhaps with subjects and allied places ; and there 
is every appearance that all four stood on a footing of perfect 
equality. Each of these cantons was sovereign, but united 
with the others by a league which was to last for ever; in 
what manner the administration of the confederacy was 
manajxed we know not. The weakness of the Samnites, in 
comparison with the Romans, arose from the fact of their not 
forming a single compact state, as the Romans did from the 
time when the Latins came under their supremacy. It was 
only in times of war that they united, though they must have 
had a permanent congress; its nature, however, is entirely 
unknown. Livy never mentions a senate of the Samnites; 
but Dionysius in his fragments speaks of their irpo^ovKoi. 
They were probably the envoys of each tribe, perhaps similar 
to the uTTOK'X'nTOi of the Aetolians; but whether these envoys 
had the right to decide upon peace and war, or whether a 
popular assembly met for that purpose, as in the states of 
Greece, is uncertain; if, however, the latter was the case, each 
tribe had a vote, for in voting the ancients never paid any 
regard to the accidental number of individuals^ belonging to 
a tribe. 

' This ob.servation removes the difficulty, which woiihl otherwise arise, in 
explaining how the majority could decide a point in an assembly in which only 


Latium received non-Latins into its confederacy; and in like 
manner Rome formed two new tribes out of the allied Vol- 
scians who lived near the Pontine marshes. At that time, 
therefore, Rome and Latium still acted in concord, each 
admitting a portion of the Volscians into its own confederacy 
and keeping the Hernicans apart. Now the relation existing 
among the Samnites was similar to that between Rome, 
Latium and the Hernicans, who were united, without any one 
of them having the supremacy, and had their common meetings; 
each of the Samnite peoples was sovereign, and united with 
the others only in regard to foreign countries. Nations which 
are threatened with destruction from without, scarcely ever 
rise to the healthy view that they must sacrifice the wishes of 
their separate elements in order to preserve their nationality: 
the people of Greece joining the Achaean league is the only 
instance of the true policy. At first the Romans and Samnites 
fought under equal circiimstances, but the Samnites never saw 
the fundamental error of their constitution. I have not the 
least doubt, that if they had reformed their constitution, 
and had instituted one senate and a popular assembly, the 
whole war would have taken a different turn. But as it was, 
the supreme command belonged to different cantons at different 
times; sometimes a measure was carried by Bovianum, some- 
times by the Pentrians, and sometimes by the Caudines : now 
one people was attacked, then another; the chief command 
passed from one people to another, and was probably given to 
the canton which was most threatened at the time, in order that 
it might be able to protect itself. The supreme magistrate of 
the confederacy bore the title of Embrntur (Impernto?'), w^hich 
is often mentioned in inscriptions. It is probable that each 
canton also had its i7npe7-afor, and that when a tribe had the 
chief command, its imperator became the imperator or perhaps 
praetor, of the whole army. There is every appearance that 
their constitutions were thoroughly democratical, as might be 
expected among such mountaineers. They must have been 


those persons voted who chanced to be present. Let us apply this to Rome : 
how was it that those who belonged to the very distant tribus Velina did not 
feel themselves wronged in comparison with the Palatina? The difficulty i 
removed, if we remember that each tribe had only one vote, so that on impor- 
tant emergencies the distant tribes sent their best men to the city, whereby the 
government became a representative one. — N. 

VOL. I. X 


completely amalgamated witli the ancient population, since, 
even after the most fearful defeats, they always appear in large 
numbers and perfect harmony. 

The extension of the Samnites towards the Liris was the 
circumstance which in A.U. 412 involved them in a war with 
the Romans. The Volscians were of no consequence: their 
power was broken, and they were for the most part allied or 
united with the Latins. The sway of the Samnites extended 
as far as Casinum, and they had subdued the Volscians as far 
as Sora and Fregellae, though sometimes they evacuated those 
districts. But they had also spread as far as Apulia, and con- 
quered a great part of that country, as for instance, Luceria. 
We thus see that they were a nation greater than the Romans 
and Latins put together, and that their country was equal 
in extent to half of the modern Switzerland. I have already 
mentioned their alliance or treaty with Rome at the beginning 
of the fifth century; but unfortunately such treaties are 
observed only so long as ambition and the love of conquest are 
not much excited. I have no doubt that the two nations had 
agreed not to extend their power beyond the Liris; but the 
Romans may have repented that they had fixed such narrow 
boundaries for themselves. Had the Samnites taken Teanum, 
they would have been masters of all the districts between the 
rivers, and have subdued the country as far as the Liris. Livy 
himself admits that the Romans had no right to form an 
alliance with the Campanians. 

It is said that the Campanians became involved in war 
with the Samnites, because the latter attacked the Sidicines of 
Teanum": the Sidicines probably belonged to the same race as 
the Oscans; they inhabited Teanum, but were perhaps not 
confined to that town. They first applied to the Campanians, 
because the latter were no longer the allies of the Samnites, 
and because the Campanian plebes could not^but consider it an 
advantage to gain the Sidicines as a protection against the 
Samnites in the north. Capua ruled over a number of towns 
all of which are said to have been Etruscan, though this is 
improbable; the territory over which its dominion extended 

' The war between the Samnites and Sidicines sliows that the dominion of 
the Samnites then extended as far as the upper Liris, so that its boundary in 
D'Anville is too narrow, — N. 


was called Campania' which was not the designation of the 
conntry which bears that name in our maps : it extended only 
a little beyond the Vulturnus as far as Casilinuni' in the south, 
and Calatia and Saticula in the north ; Nola, Neapolis, Pom- 
peii, and Herculaneum did not belong to it; the territory 
therefore was small, and the name denotes only the domain of 
the citizens of Capua. In consequence of the fertility of their 
country, the Campanians were wealthy and unwarlike; they 
were anxious to prevent the attack of the Samnites, but being 
unable to resist the mountaineers they were defeated. The 
Samnites proceed to Mount Tifata, above Capua, and laid waste 
the country all around. It was the ancient Oscan population 
of Capua that carried on the war in spite of the Samnite 
colony : their distress was very great, and it is likely that the 
Samnites contemplated restoring the oligarchical constitution 
of the colony: under these circumstances, the Campanians 
applied to Eome, or probably to the diet of the Romans, 
Latins, and Hernicans. This is evident, from statements 
derived from L. Cincius; in Livy we perceive the intentional 
obscurity of the Roman tradition about it. The Romans 
themselves would have been greatly perplexed by this proposal, 
as they were allied by treaty with the Samnites ; hence the 
Campanians placed themselves under the protection of the 
whole confederacy. This deditio must not be imagined to be 
that of a conquered people; for here we merely have one 
nation which seeks protection, and another which grants it. 
In such things, the Romans were always hypocritical observers 
of the letter of the law, though in reality they might act in 
direct opposition to the spirit of the laws of Numa and Ancus ; 
the only good result of this feeling, was, that they always 
wished to have at least the appearance of justice on their side. 
We must not, however, on this account, consider the ancient 
Roman fides as altogether hypocritical, since their reverence 
for law certainly did keep them from many an act of oppres- 
sion towards the weak. They may be excused by the con- 
sideration that according to all appearance the Samnites were 
becoming too great; it could be foreseen that, after all, the 
treaty would soon be violated, and hence they would not allow 

^ Campania is the country of the Campanians, that is, the inhabitants of 
Capua. On coins wc read Capani, and in Plautus we find Campas instead of 
Campanus. — N. 

x 2 


a favorable opportunity to pass by. The Romans, however, 
were too much tempted by the prospect of gaining the 
Campanians and all the people of that country by forming a 
treaty of protection with Capua. There is no question that 
they were not impelled by a desire to protect those w^ho were 
in want of aid; they were overpowered by an evil spirit, and 
the exasperation of the Samnites against them was perfectly 
just. The Romans sent an embassy to the Samnites, requesting 
them to conclude peace with the Sidicines, and not to lay 
waste the Campanian territory because Campania had placed 
itself under their protection. The Samnites proudly rejected 
this proposal; and now arose their gigantic struggle against 
the Romans, Latins, and Hernicans. 

This Samnite war is the first in Roman history that is 
worthy of being related; whatever deduction we may make 
irom the numbers stated by Livy, — which we may do the 
more safely, as the person of whom these deeds are narrated 
is a Valerius, and Valerius Antiat was a client of that Hxmily 
— yet the difFerenco between these battles, and the earlier 
ones is obvious. In the year A.u. 412, three battles were 
fought, the first great battles, excepting that of A. Postumius 
Tubertus, on IMount Algidus, that are recorded in Roman 

In this year the Licinian law was violated for the last time: 
both consuls were patricians, A. Cornelius Cossus, of whom but 
little is known, and M. Valerius Corvus, a man in whose favour 
an exception might have been made at any time. He was, as 
Pliny justly remarks, one of the greatest and happiest men, 
and Solon himself would have admitted it. He is one of the 
historical heroes of Rome, although the story about the origin 
of his surname belongs to poetry (Livy himself does not con- 
sider it histoi'ical): but it proves, that even as late as that time 
the heroes of Rome were the themes of sonsf. No one will 
believe that in A.U. 406, a Gaul challenged the boldest Roman 
to a single combat, and that Valerius, then only twenty-three 
years old, conquered him, a raven flying against the enemy, 
and pecking at and tearing his face, so as to render the victory 
easy for the youth. His first consulship falls in his twenty- 
third year, the one in which he had slain the (Jaul: it is pro- 
bable that forty-six years later he was raised to his sixth consul- 
ship: he lived to nearly the age of one hundred years, and 


saw the complete subjugation of Italy. At that time it was 
still a matter of frequeut occurrence, that men, after their 
consulship, were invested with the other curule magistracies; 
to these Valerius was repeatedly elected down to his latest days, 
and discharged the duties of all with the full vii'our of his 
mind. He is the man who may give his name to the century 
he lived in ; he was the idol of his soldiers, being not only one 
of the greatest generals, but swaying the hearts of his soldiers 
by his amiable and brotherly manners, without ever losing his 
authority over them : the soldiers saw in him the ablest of 
their equals. If we imagine ourselves placed by the side of 
his death-bed, and look back upon his life, full of important 
events, we shall have before our minds' eye a gigantic period 
which we cannot picture to ourselves with too much distinct- 

Rome sent two consular armies, one-half consistinf of 
Romans, and the other of Latins, into Campania, which on 
the side of Samnium was quite open. Nola was even a Sani- 
nite colony, and Keapolis was allied with them. The two 
armies appear in entirely different circumstances. That of 
J\I. Valerius was in Campania, beyond the Vulturnus, and 
acted evidently quite on the defensive. The army of Corne- 
lius Cossus, on the other hand, was destined to make a 
diversion into Samnium, Capua undoubtedly being the basis 
of that operation, since he penetrated into Samnium to the 
north of the Vulturnus, by the common road from Calatia to 
Beneventum. We cannot obtain a clear view of the events 
of the war, and can judge of their course only by drawing 
inferences from isolated facts. We find Valerius on Mount 
Gaurus, probably near Xuceria, so that the Romans entered 
Samnium on that side for the purpose of protecting Campania. 
There was another ]\Iount Gaurus, not far from Cumae and 
Cape ]\Iiseuum. If the latter is meant, the Romans must 
have been pressed by the Samnites into that corner, and 
having the sea and the Vulturnus in their rear, their victory 
would have been the result of despair.^ This would clearly 
show, that at first the Romans sustained losses which are 
passed over by Livy, or the annalists whom he followed ; but, 

* In his Hist, of Rome, in. p. 119, Niebiihr speaks with much more coufidenee 
in favom- of tlic second view; but it must be observed tli;U that passage, wiiii 
the same words, occiu's also in the first cditiou (181:2), whereas tlie opinion 


at all events, the battle restored the balance. It was obviously 
the greatest of all that had yet been fought by the Romans, 
for though previous battles may have been bloody, yet they 
were not carried on with perseverance. AVhen the Gauls had 
fourrht for a few hours, and to no purpose, they gave up the 
battle ; and the Aequians, Volscians, and Hernicans were few 
in number. The Samnites, on the other hand, were arrayed 
against the Romans in equal numbers, and possessed equal 
determination, and thus they fought the whole day till night- 
fall without any decisive result, until the Roman equites, the 
j)rincipes juventutis (the Samnites had no cavalry, and that of 
the Romans was weak), dismounted, placed themselves before 
the lines, and fought with true heroism. The real nobility 
of the nation put all the rest to shame, but the latter now 
followed their leaders and were irresistible. The massacre 
was immense on both sides; the Samnites yielded, but only 
retreated. It was not a flight, but just as at Grossgorschen 
and Bautzen ; and the conquerors followed them with the 
greatest caution. In the neighbourhood of Suessula, only a 
few miles from the battle-field, the Samnites made a fresh 
stand. Their camp and the wounded, of course, fell into the 
hands of the Romans. The victory gave to the latter more 
hopes than real advantages; but the main point was, that the 
battle was a happy omen for the whole war, which they had 
certamly begun with the prospect of a possibility of their 
being in the end completely annihilated. 

The expedition of A. Cornelius Cossus into Samnium 
luidoubtedly belongs to the beginning of the campaign. He 
seems to have been met by a general rise of the militia of the 
Samnites, whose general custom it was to act on the offensive 
with the army, and to leave their country to the defence of 
thq people: the invading Romans had mostly to do with the 
country people who rose in arms. Samnium was then in full 
vigour and strength; the Roman commander incautiously 
entered the hostile country, which was unknown to him and 
very difficult to pass with an army. No enemy opposing him, 
he crossed from west to east the chain of mountains which 

expressed in our text, is that given liy N. in his lectures of 1828-9. The 
detailed description of the battle, liowcver, belonging to the year 1S26, is based 
upon the explanation which he had given at an earlier period. Wc make this 
observation to prevent misconceptions. En. 

p. DECIUS MUS. 311 

runs from north to south. There Tvere only a few passes : the 
first cokunn was ah'eaJj in the valley, wliile the rear was yet 
on the ridge of the mountam — this is what we can gather 
from the confused accoimt of Livy. The consul's intention 
probably was to reach the road to Beneventum and the fertile 
valley of the Galore, in order to separate the northern from 
the southern Samnites. When in this situation, he observed 
that the opposite hill was occupied, he halted: to retreat 
through the defile was very difficult, and the Samnites were 
advancing to occupy a height commanding the road. The 
Romans were almost surrounded, for the Samnites were 
already engaged in taking possession of the road in their rear. 
AVhile the Romans were thus situated, the tribune, P. Decius 
Mus, who belonged to one of the greatest plebeian families, 
offered to the consul to hasten iip the mountain with one 
cohort, and to take possession of the height which the Sam- 
nites incautiously had just abandoned, so that he could attack 
them in their rear, and sustain the shock of the enemy, until 
the Roman army should reach the ridge of the mountain through 
the pass. This plan was carried into efiect. Decius reached 
the height which commanded the pass, before the Samnites, 
who were now obliged to try to dispossess him ; but he there 
fought with his men like the Spartans at Thermopylae, in the 
conviction that they must die, and with such perseverance, that 
the Samnites gave up the attack for that evening. ^Vhile the 
Romans retreated to the road which had been abandoned, the 
Samnites encamped with the determination to storm the height 
the next morning. The battalion of Decius, however, was 
quite surrounded; but in the night he ventured to sally down 
the hill, and forced his way through the enemy, and thus with 
the survivors of his band he returned to the consul. It is stated, 
indeed, that on the day following the Romans again won a 
great victory, but we cannot trust the account. The army of 
Cossus is not after this time mentioned : he had probably be- 
come aware of the perilous nature of his expedition, or he was 
called into Campania, because some loss had been sustained 
there. On ]\Iount Gaurus, Valerius was alone, but at Suessula 
we find the two consuls united : those enemies who followed 
the march of Gossus joined the Samnites. Both were encamped 
opposite each other for a long time, but the Samnites being 
superior in numbers, considered their cause too safe. Their 


commander cannot have been a man of much talent; they 
ranged over the country indulging in plunder, especially as 
Valerius in his fortified camp seemed to show symptoms of fear. 
When the Samnitcs were thus scattered, Valerius suddenly at- 
tacked their camp and took it ; he then quickly turned against 
the separate corps and routed them one after another, so that 
both consuls gained a brilliant victory and were honoured with 
a triumph. 


The Romans now experienced that times may be bright and 
prosperous, although a heavy pressure is weighing upon the 
people. Ever since the Licinian law, the misery of the mass 
of the citizens continued uninterrupted, and ever and anon new 
commissions were appointed to liquidate the debts, but without 
any good result. The wars demanded very heavy taxes, and 
the plebeians were obliged to fight in the battles, and at the 
same time to provide for their families : we have reason to 
believe tjiat not half the men capable of bearing arms remain- 
ed at home ; and so bloody a war as that against the Samnites 
must have caused severe sufferings to many a family. In the 
second year of the war, when either the Latins had the su- 
preme command, or, perhaps, a truce existed between the 
Romans and Samnites, a mutiny arose which very nearly came 
to an insurrection. Livy is obscure on this subject, but an excerpt 
of Constantino from Appian, in which we distinctly recognise 
Dionysius, throws much light upon it. The insurrection of the 
year A.u. 413, was brought about by the state of the debtors; 
Livy conceals this, and relates, that, while the Roman army 
was encamped in Campania, probably in consequence of a 
truce, the soldiers were tempted to make themselves masters of 
Capua. The Roman consul who undertook the command and 
found the army engaged in a manifest conspiracy, endeavoured 
to get rid of the ringleaders by sending them one by one in 
different directions, and then ordering each to be arrested. 
This mode of acting however excited their suspicions, and one 


cohort which was sent to Kome haked near Lautulae, between 
Terracina and Fundi, four or five miles from the former, in a 
desolate district between the hills and the sea, which was at 
all times the haimt of robbers and banditti. The hills there 
approach the sea almost as near as at Thermopylae, though 
they are not so steep: it is quite a narrow pass by which 
Latium and Campania are connected. There seem to have 
been warm springs in this place, so that even in the name 
there is a resemblance between it and Thermopylae. The 
country is now desolate, and when I was at Terracina I forgot 
to enquire for the springs, in consequense of which I was not 
able to find them. In the second Samnite war a battle was 
fought near Lautulae, which is one of the greatest battles re- 
corded in history. When the cohort reached that place, it 
revolted and was joined by a number of others; the communi- 
cation between Eome and the head quarters of the army was 
cut off; the messengers of the consuls were intercepted, and 
we must suppose that the whole army refused to obey its com- 
manders. A number of persons who v.'ere enslaved for debt 
attached themselves to the insurgents, and what now happened 
was more terrible than any thing which Rome had yet ex- 
perienced, for the insurrection became general, and the common 
people marched against the city in arms, though they did not 
injure the consul. This multitude was no longer the plebes of 
the Sacred ]\Iount : it was an insurrection of the proletarians 
against the rich, and very like a revolt of the workmen in a 
factory against their employers. But fortunately for Rome, 
they were not yet quite impoverished : they still looked upon 
themselves as plebeians, and upon the most distinguished 
among the plebeians as their leaders, so that the latter might 
make use of them in reforming the constitution. It is surpris- 
ing to find that they fetched T. Quinctius, a lame and aged 
patrician, from his estate in the territory of Alba, and made 
him their captain, just as the peasants in the peasant- war of 
Germany made GiJtz their leader: they then advanced towards 
the city which was thrown into great alarm l)v the approach- 
ing danger. The government no longer knew in whom to trust : 
everybody in the city armed himself as well as he could ; but the 
civic legions would scarcely have been able to maintain them- 
selves against the arm v. The heart of Valerius Corvus was 
bleeding at the prospect of a civil war; tlic plcbs too was 


fortunately not quite demoralised; and he made an offer of 
reconciliation. The army likewise was moved; when they 
saw their relations armed in the city, they raised loud lamenta- 
tions and were wiUing to listen to proposals of peace : both 
parties were loth to shed the blood of their brethren. The 
consequence of this moderation on both sides was a reconcilia- 
tion, and a peace was concluded in which, according to Appian, 
that is, Dionysius, the debts were cancelled. 

The cause of the insurrection, as it is described in this 
account, is in the highest degree improbable; the sending away 
of individuals could surely have lasted only a very short time, 
and it is quite inconceivable that a whole cohort should have 
been thus disposed of. The other account does not speak at 
all of an insurrection of the soldiers, nor of their intention to 
take Capua, but represents it as an internal commotion, as a 
secession, like those of the commonalty in former times, and 
as having arisen out of the distress of the numerous debtors, 
and the disputes between the patricians and plebeians, since 
the Licinian law had ceased to be observed. The plebeians 
seceded to the neighbourhood of Alba, where they were joined 
by cohorts from the army. The senate, it is said, levied troops, 
but there is no mention of the two armies having met, nor of 
the appointment of Valerius to the dictatorship of which Livy 
speaks : when matters had proceeded so far as to call for the 
interference of the sword, both parties determined to put a 
stop to the dispute at any cost. 

These occurrences were followed by a great and essentially 
civic legislation, by which that of Licinius was completed. 
Whatever may have been the real history of this commotion, 
it must, at all events, have been more important than Livy^s 
description would lead us to suppose. Up to that time the 
Licinian law respecting the consulship had been violated seven 
times in the course of thirteen years; but henceforward we 
hear of no more actual violations, although some absurd attempts 
still continued to be made. During that commotion some 
regulation must have been made which rendered it impossible 
for any attempts against the Licinian law to succeed; and 
clauses must have been added, perhaps as severe as those in the 
lex Valeria Horatia, by which the heaviest penalties were 
inflicted on him who shoidd disturb the election of the tribunes 
of the people. It is, moreover, said to have been determined 


that botli consuls might be elected from among the plebeians, 
but this seems to be a misunderstanding, and it can at any rate 
be proved that no such regulation was carried into effect. In 
the Hannibalian war, a special resolution was once passed that 
during the war both consuls might be elected from among the 
plebeians, but no practical application of this resolution was 
made, and it was not till the year A.u. 500, that the natural 
principle gained the upper hand; the patrician nobility had 
then become so insignificant, that it was impossible any longer 
to abide by the law of Licinius. 

Another regulation mentioned by Livy is of great impor- 
tance, and shews that the question was no longer merely about 
the difference between the two estates, but that the plebeian 
nobles had begun to have recourse to the oligarchical intrigues, 
which until then had been employed by the patricians alone, a 
proof that the one set of men was not better than the other. 
The law in question established two points, first that no one 
should hold two curule offices at the same time, and secondly 
that a person invested with a curule office should not be re- 
eligible to it till the expiration of ten years. The first provision 
could affect only the patricians in regard to the praetorship, 
and was probably made because it had often happened that a 
patrician consul had at the same time caused himself to be 
elected praetor, in order to obtain an influence over his col- 
league; in regard to the aedileship, it may have affected 
the plebeians also in alternate years. Livy says that the law 
was mainly directed against the ambitio novorum hominum; 
the second provision of the law had probably been brought 
about by the plebeians themselves, as a security against the 
overwhelming influence of members of their own order, for 
up to that time we always find the same plebeian names in the 
consulship, such as Popilius Laenas, C. Marcius, C. Poctelius, 
so that it was intended to prevent the exclusive lustre of a 
few plebeian families. 

Livy was aware of the existence of two laws respecting 
military affairs which arose out of this insurrection. The first 
enacted that whoever had once been a tribune of the soldiers 
should not afterwards be made a centurion : it is said, that this 
law was enacted through the agency of a certain Salonius who 
had been thus degraded by his enemies. The consuls had it 
in their power to appoint the centurion? : when a person had 


been tribune, it was contrary to the feeling of the soldiers 
that he sliould become a centurion, because a centurion was 
only a subaltern officer. Six of the tribunes were annually 
appointed by the tribes, and the rest by the consuls, but a 
person could not be elected for two successive years by the 
same authorities. During the year in which he could not be 
tribune, he would be free from military service. Now Salonius 
who had been tribune^ and in that capacity seems to have been 
in opposition to the consuls was appointed centurion by them : 
the consuls thus degraded him while he was raised by public 
opinion, and it was against such proceedings that the law was 
directed. The regulations about the corps of officers are among 
the most excellent adopted by the Romans. Slow and gradual 
advancement and a provision for officers in their old age were 
things unknown to the Romans. No one could by law have a 
permanent appointment; every one had to give evidence of his 
ability ; the idea of a gradual rising from the ranks and of a 
standin o- corps of officers was never entertained : a tribune of 
the soldiers was elected for one year only, and if he shewed no 
skill he was not re-elected ; but he who was fit was elected 
year after year, sometimes by the people, sometimes by the 
consuls. It was, moreover, not necessary to pass through a 
long series of suljordinatc offices; a young Roman noble served 
as eques, and the consul had in his cohort the most distin- 
guished to act as his staff; there they learned enough, and in 
a few years a young man, in the full vigour of life, might 
become a tribune of the soldiers. But besides this, due atten- 
tion was paid to that respectable class of people who without 
talent for higher posts were well fitted to train the soldiers. 
Such persons became centurions, who may be compared to our 
sergeants ; all of them were people of common descent, they 
had good pay and a respectable position, and in special cases 
Avhere a man shewed particular ability he might become 
tribune. All the functions which, in modern armies, are 
performed by a large number of subaltern officers, might just 
as well be performed by an able sergeant. The military regu- 
lations of the Romans in all these points, are as admirable as 
those concerning the training of the individual soldier. 

The second law shews how Livy confounds everything: the 
pay of the cquites is said to have been diminished because 
they had taken no part in the insurrection. If the mutineers 


could have carried siich a law, the state would have been lost. 
I believe that this was the time when the equitcs ceased to 
receive their 2000 asses from widows and orphans, and wlien 
it was established that they should have a fixed pay, — a fair 
change, but a disadvantage to the eques publicus ; fair, I say, 
because the state was able to bear the expense. 

The curies, assembled in the Petelinian grove, now decreed 
a full amnesty for all that occurred, and no one was to be 
upbraided, either in joke or in earnest for his conduct. Livy 
considers it to have been a decree of the centuries, audoribus 
patribus, but from the trial of ]\Ianlius it is clear, that only the 
curies assembled in the Petelinian grove. 

Hereupon the Romans concluded peace Avith the Samnites : 
even the year before, they had received from them a compen- 
sation for pay and provisions, or they received it now. The 
peace was concluded by the Romans alone, and that with a bad 
intention, for they had undertaken the war conjointly with 
the Latins, whom they now left to shift for themselves. They 
gave up Capua to the Samnites, and left the conquest of 
Teanum to their choice, but the Sidicines threw themselves 
into the arms of the Latins, and concluded a separate alliance 
with the Volscians, Aunmcans, and Campanians. Such things 
have occurred in modern times also, as for example, the alliance 
between Prussia and Russia under Frederick the Great, and 
Peter the third, in the Seven Years' War. The Latins now 
continued the war, suo Marfe, whicli Liv}^ in accordance with 
his peculiar views, regards as an act of injustice on their part, 
as if they had thereby offended against the majesty of the 
Roman people. They made war upon the Pelignians, from 
which we see that the Aequians belonged to them, since other- 
wise they could not have come in contact with the Pelignians : 
the latter allied themselves with the Samnites, and the Samnites 
required the Romans either to act as mediators, or to give 
them succour; for the peace with them had immediately been 
followed by an alliance. The alliance of Rome with the Latins 
and Hernicans had now come to a crisis; the Plernicans were 
either neutral, or, as is more probable, were still allied with 
the Romans, since Livy and the Capitoline Fasti do not mention 
them among those who triumphed over i\Iaenius. Such con- 
federacies may exist among nations, none of which is as 
ambitious and powerful as Rome then was: but now three 


things were possible ; they might cither separate and remain 
friends, or form a union like that of Great Britain and Ireland, 
or lastly, decide by force of arms which was the strongest; for 
to continue together, side by side, was impossible. Even the 
year before, the war had not been carried on in common, and 
the Latins had gone into the field under their own banners. 
Hence it was now resolved to negotiate. The Latins had a 
more solid constitution than the Samnites, and were governed 
like the Romans; they had two praetors as the Eomans had 
two consuls; and they must have had a senate, since there is 
mention of the decern primi, who were evidently the deputies 
of so many towns. These decern prirni went to Rome, and 
there made the very fair proposal that the two states should 
unite. The senate was to be raised from 300 to 600; the 
pojDular assembly was to be increased (so that it would probably 
have been necessary to increase the twenty-seven Roman tribes 
to thirty, and to allow the Latin towns to vote as so many 
tribes), Rome was to be the seat of the government ; and every 
year one of the consuls was to be a Roman, and the other a 
Latin. Had the Romans accepted this proposal, Rome and 
Latium would in reality have become equal ; but every Roman 
would have felt his own infiuence weakened. A Latin consul 
was repugnant to the Romans; for in all republics, however 
democratical they may be, there is a spirit of exclusiveness, of 
which we have a striking example in the history of Geneva, 
where we find citoyens, bourgeois, natifs, that is, the children of 
the metoeci or habitans, and lastly habitans, all of which classes 
acquired the franchise one after another. The canton of Uri 
is the most oligarchical of all. The plebeians as well as the 
patricians were indignant at the proposal ; as there was to be 
only one Roman consul, the question would have arisen, is he 
to be a patrician or a plebeian? they would more easily have 
adopted a proposal to have four consuls. The embassy of the 
Latins, as Livy says, was received with general indignation, 
not because the Romans were ignorant that the impending 
struggle would be a contest for life and death, but because 
vanity and selfishness outweighed this consideration. It is 
related that the consul, T. ]\Ianlius, declared that he would cut 
down with his own hand the first Latin who should appear in 
the Roman senate. The story moreover has this poetical addi- 
tion, that while the discussions were going on in the Capitol, 


tliere arose a tempest, accompanied by a heavy fall of rain, and 
that the Latin praetor, in descending the hundred steps of the 
Tarpeian rock, fell down, and was picked up a corpse; the 
unpoetical spirit of later narratives has changed his death into 
a fit of fainting. 

The Sabines with their ancient reputation for justice had 
sunk into a torpor and had lost all importance; the northern 
confederacy of the Marsians, Pelignians, Marrucinians and Ves- 
tinians, brave as they were, had no other wish than to live 
quietly among their mountains. The Komans were allowed to 
march through their territory, and as they were allied with 
the Samnites, the latter expected that the Latin war would 
aiford them an opportunity of taking Capua and Teanum. If 
the Romans had dreaded to allow their territory to be ravaged 
by the Latins, they would have been obliged to maintain 
themselves on the defensive, or to carry on tedious sieges of 
the Latin towns. But the Roman commanders here shewed 
their greatness : they formed a most masterly plan, made up 
their minds to the very boldest undertakings, called out the 
army of reserve in the city, and abandoned their territory up 
to the very gates of Rome to the Latins. Their army marched 
through the Sabine and Marsian territory, taking a circuitous 
route in order to join the Samnites, and in conjunction with 
them proceeded towards Capua. If the Latins had abandoned 
the Campanians to their fate, and had gone to meet the Ro- 
mans on their march, in the country of the Aequians, they 
might perhaps have defeated them in those impassable districts. 
But a great general places his enemy where he wishes to have 
him : the daring boldness of the Romans is a proof of the 
excellence of their generals, Manlius and Decius, who, like 
all great generals, had formed a correct estimate of their ene- 
mies, and trusting to the accuracy of their estimate, ventured 
to lead their army by that circuitous road. Had the Latins 
moved rapidly, they might have laid waste the whole Roman 
territory, appeared at the gates of Rome eight days before the 
Roman army could have returned, and efiected an easy retreat 
to their fortresses ; but the Roman generals probably knew 
that their enemies were timid and without great leaders, and 
therefore left the road to Rome unprotected. The Latins 
listened to the complaints of the Campanians, and perhaps 
imagined that the Roman army might be annihilated then at 


one blow, since it could not return. Their forces justified them 
in this expectation, and the decision of the contest hung upon 
a thread ; for there was as much probability of their conquering 
as of being conquered. The Romans undoubtedly had enlisted 
all the men they could muster, but they were, notwithstanding, 
inferior to the Latins in numbers: it is quite certain that they 
were joined by the Samnites, though the Roman annals endea- 
voured to conceal that fact by stating that the Samnites did 
not arrive till after the battle. The Latins and their allies, 
the Volscians, Aequians, Sidicines, Campanians and Aurun- 
cans were encamped on the eastern side of Mount Vesuvius ; it 
is vmcertain, whether Veseris, where the battle was fought, is 
the name of a place or of a river. The two armies faced each 
other for a long time, dreading the day which was to decide 
their fate. If the Latins had had an able commander they 
might, even after a defeat, have been better off than the Romans, 
as they might have withdrawn to Capvia, and protected them- 
selves behind the Vulturnus and Liris, and there collected re- 
inforcements from their own country. The Romans, moreover, 
in a military point of view were not superior to the Latins; 
one Roman and one Latin century had always formed a mani- 
ple in the Roman legion, so that the constitution of the two 
armies was the same. Under these circumstances, the consul 
forbade, under penalty of death, all skirmishes, on account of 
the importance of the moral impression that might thence result, 
trifling events easily producing a prejudice regarding the issue 
of a battle, and not on account of the acquaintance of the Ro- 
man soldiers with the enemy, as Livy states. In like manner, 
it was forbidden in the Russian army to accept the challenge 
of the Turkish spahis. The stricter the command was, the 
more did the Latin horsemen provoke the Romans^, and this 
gave rise to the single combat between Geminius Metius of 
Tusculum and the son of the consul IManlius. This occurrence 
is beautifully described by Livy, with the heart of a Roman 
and the power of a poet : the father in order to enforce obedi- 
ence ordered his unfortunate but heroic son to be put to death. 
There is yet another circumstance which Livy mentions but 
cursorily": there can be no doubt that in the ancient tradition 

' The Roman cavalry was always tlie worst part of the army, and inferior, for 
example, to that of the -i'Etoliaiis. — N- 
' VIII. 8; towards tlie end. — Ed. 


there was, besides IManlius, a centurion who gained the victory 
for the pedites, as the son of the consul did for the equites. 

The long time which passed away before the battle began is 
a decided proof that the Samnites joined the Romans. Both 
parties commenced the fight with sail forebodings; and the two 
consuls, moreover, had had a vision prophesying a disastrous 
issue by informing them that one army and the commander of 
the other were forfeited to the gods of the dead : the two con- 
suls therefore agreed that the commander of that wing', which 
should first be in danger should devote himself to the infernal 
gods. Each of them oflfered up a sacrifice : that of Decius was 
unfavorable, but that of Manlius promised success. It is men- 
tioned on this occasion, as on many others relating to sacrifices, 
that the liver had no caput : the caput is the same thing as in 
Italian is still called capo, that is, the part where the liver is 
connected with the diaphragm; and the caput being wanting 
means that there was no trace of the connection. The liver 
presents the greatest varieties, even in animals which are per- 
fectly healthy. The heart and lungs afford no means for form- 
ing prophecies, while the liver in almost every case has some 
abnormity. Decius, then, went into battle with the intention 
of sacrificing himself; but the resolution must have been made 
even before he left Rome, since the pontifex accompanied him 
for the purpose of dedicating him to the gods. 

The Roman legion at that time consisted of five battalions, 
hastati, principes, friarii, rorarii and accensi. Among them 
were three battalions of tlie line, mixed with light-armed troops, 
and one battalion of light troops, the rorarii with one third of 
the hastati. Nearly two thirds of the hastati had, from the 
earliest times, been armed with lances; the p7-incipes had pila 
as early as the time we are here speaking of, but the triarii still 
had lances. These formed the troops of the line ; but the feren- 
tarii were light troops with slings, and one third of the hastati 
also were light troops armed with javelins. They were placed 
in front at the beginning of a battle, just like the ■^ikoi of the 
Greeks, and afterwai'ds withdrew through the lines, and placed 

^ It is a general mistake of modern writers to compare the cornu dextrum and 
sinistrum with aiTangements of our own armies, and consequently to suppose 
that there also existed a central battalion (corps de bataille^; but a Roman .nrmy 
consisted only of those two halves (cornua). All modern writers on tactics, with 
the exception of Gui^chard, are mistaken on this head. — N. 

VOL. I. Y 


themselves behind them, but always advanced again as soon as 
the enemy retreated. These three battalions stood in single 
maniples with intervals, as at Zama, but cannot possibly have 
been drawn up en echelons, since so large an interval in one 
line as that described by Livy is practically impossible, for the 
cavalry would immediately have broken through it ; they were 
probably drawn up in the form of a quincunx, and in this 
manner the intervals may be conceived. Now as all the Roman 
military arrangements were calculated to support the efforts of 
individuals as long as possible, and not to form solid masses like 
those of the Greeks, the first two battalions were drawn up as 
near as possible to the enemy and under the protection of the 
light troops. Every Koman soldier was perfectly trained in 
the art of fighting. According to later regulations, the soldiers 
began with the pilum. The Roman soldiers were drawn up 
in ten lines with large intervals, and when they were drawn up 
close, the first battalion advanced, stopped and then threw the 
fearful pila, which penetrated through the coat of mail, and of 
which each soldier had several. After the first charge, the 
soldiers who had first thrown the pila retreated two steps, while 
those who stood behind them advanced two steps, and occupied 
places in the line by their side ; the first line then withdrew 
and formed the tenth line, and thus all the ten lines had their 
turn for making use of their pila. This mode of attack, which is 
the only true and possible one, was formidable for the enemy. 
If we consider this quiet mode of advancing and retreating, we 
can understand why the battles lasted so long, and why the 
combatants did not at once come to close quarters ; one hour 
must undoubtedly have elapsed before all the pila were thrown, 
and then the fight with swords began, during which the lines 
again took it in turns: those who stood behind were not idle, 
for when the foremost fell or were fatigued, those in their rear 
advanced and took their places ; and thus a Roman battle might 
have lasted a long time. To fight successfully in such a battle 
the soldiers must be trained and drilled in the excellent manner 
of the Romans: the dust and the war-cries were not disturbing 
as smoke and the thuntlcr of cannon. When the hastati had 
done fighting, they withdrew behind the principcs who then 
commenced : when they were overpowered, they fell back upon 
the triarii, who at that time formed a kind of reserve, which, 
however, was obliged to take part in the fight. Besides these 


four battalions, the three battalions of the line, and tlie one 
with light armour, there existed a fifth consisting of the acccnsi 
who were Avithout armour, and whose business it was to step 
in and take the armour of those who had fallen; the accensi 
and velati were the two centuries that were added to the fifth 
class, but did not come up to its census. It is clear that IMan- 
lius in that war did something which had never been done 
before : he armed the accensi, made use of them instead of the 
triarii to strengthen the lines, and reserved the triarii for the 
decisive moment, and by this means he saved himself. Livy 
states that the Latins mistook the accensi for the triarii, which 
is impossible; but the accensi likewise may have been armed 
with spears and have advanced as phalangitc?. The Latins 
followed their old routine, and their battle-line consisted of the 
most ordinary elements. The wing commanded by Decius 
fought without success and the Latins conquered, whereupon 
Decius ordered himself to be devoted to death by the pontiff 
M.Valerius. This devotion inspired the whole army with 
fresh courage and was at the same time believed to have a 
magic effect upon it, since the consul had atoned for the whole 
nation, which was now considered invincible. Hence, as tra- 
dition states, fate turned all at once : the legions rallied and 
gained the most complete victory. 


If Rome had succumbed in this war, the whole Roman army 
would have been annihilated; but the Latins could not have 
derived the same advantages from their victory as were gained 
by the Romans: as Latium itself had no unity and was with- 
out a great central point, the sovereignty of Italy would have 
been undecided between it and Samnium. There is every 
probability that Italy would then have fallen vmder the domi- 
nion of foreigners ; it would perhaps have become a pennanent 
prey of Pyrrhus or of the Carthaginians, and the Gauls would 
have ravaged it incessantly. Had the Italian nations been 
wise, the same state of things might have been developed as 



we afterwards find in existence, but it would liave taken place 
without violence and destruction. Rome conquered Italy, but 
this subjugation is nevertheless the most desirable thing that 
could have happened to Italy. 

The defeat of the Latins in the battle described in the 
previous Lecture must have been complete, and so decisive, 
that all were seized with a panic. Capua evidently submitted 
at once, and the defeated did not even attempt to protect 
themselves behind the Vulturnus, but at once retreated across 
the Liris. Notwithstanding the general flight, however, a new 
army formed itself at Vescia, an Ausonian town near the 
Vescinian hills, and probably the modern S. Agata di Goti ; 
there are indeed no ruins, but many sepulchral monuments; it 
is situated on the natural road from the Liris to the Vulturnus, 
the mo'intains being on the left of the road to Naples. The 
flight of the Latins therefore cannot have been as disorderly 
as Livy describes it. There the survivors assembled and were 
reinforced by the contingents of the ancient Latin and Volscian 
towns; the Volscians on the sea-coast and the Liris, the 
Auruncans and Sidicines, that is, the whole country between 
the Liris and Vulturnus was united, and oflTered a final battle 
to the Romans near Trifanum, on the Liris, between Sinuessa 
and Minturnae. The Romans immediately, and even before 
completing their march, attacked the enemy, and gained a 
decided victory, but with great loss : this second defeat of the 
Latins completed the destruction of all their resources, espe- 
cially as they had the broad Liris in their rear. The contingents 
dispersed to their respective homes in order to defend them. 
The Romans made immediate use of their success, and advanced 
through the territory of the Latins towards Rome. Now 
whether, as Livy relates, Latium was completely subdued as 
early as that time, or afterwards, cannot be determined with 
any certainty, for the Latins again appear as enemies in the 
following year. There are many circumstances in antiquity of 
which we can say, that they must have been such or such ; but 
this is not the case -with events which are accidental : le vrai 
vtest pas toujours vraiscvihlable. I will therefore not assert posi- 
tively whether the Latins, in their first consternation, laid 
down their arms and afterwards took them up again. But 
however this may be, the senate pronounced the sentence, and 
with lofty confidence in the certainty of success resolved that 


the ayer publicus of the Latin state, the Falernian district of 
the Campanians, and part of the ager Privernas (Frivernum 
seems not to have joined the Latin league) should be confiscated 
and assigned to the plebeians viritim, that is to every one who 
wore the toga pura ; assignments beyond the Vulturnus would 
have been of no value to the Koraans. The allotment, how- 
ever, was made on a small scale, owing to the plebeian nobles 
having intrigued with the patricians against the multitude. 
An annual revenue of 450 denarii was assigned to the Campa- 
nian eqiutes, probably as a compensation for the ager Falernus, 
and this sum had to be paid by the commonalty of Capua; it 
has already been observed that these equites consisted of the 
Samnites of the ancient colony, who anxious for their own 
interest, had not taken any part in the war. In the year 
following, the Latins again appeared in arms, probably because 
the Komans, after receiving their deditio, had driven them to 
despair by the fearfid punishment inflicted upon them. We 
know, from several examples, with what cruelty the Romans 
acted towards a revolted people, witness Pleminius at Locri, in 
the Hannibalian war ; hence we may imagine that the garrisons 
of each town were allowed every possible license, and such a 
place had to suffer all the horrors of a town taken by the sword. 
The Romans now made war against the Latins from the nearest 
points of their own territory: the insurrection existed only in 
the ancient Latium proper, at Tibur, Praeneste, and Pedum 
on the one hand, and at Aricia, Lavinium, Antium and Veli- 
trae on the other; Velitrae was originally Latin, afterwards 
Volscian, and in the end it received a Roman colony ; Tusculum 
and Ardea were Roman. These places form two masses, each 
of which endeavoured to defend itself The two consuls Ti. 
Aemilius ]\Iamercinus and Q. Publilius Philo fouirht against 
them. Publilius had frustrated an attempt of the Latins to 
maintain themselves in the field* ; Avhile Aemilius besieged 
Pedum. There the Tiburtines, united with the people of 
Pedum, had fortified themselves, and the year passed away 
without any success. For reasons which are luiknown to us, 

' In one of the MSS. we find in Campis Tincetanis instead of " in tlic fiold;" 
but this has evidently heen entered after the Lecture by a student wlio b;nl left 
a gap during the Lecture, and Nicbuhr probably al hides to thcCampi Fenectniii 
mentioned bv Livv.,— Ed. 


a dictator was now appointed ; and Acmilius took this opportu- 
nity of confemng that dignity upon his colleague Publilius. 

There now followed a cessation of hostilities, whether in 
consequence of a truce or from other causes, is utterly unknown, 
and a course of internal legislation to curtail the rights of the 
patricians engrossed every body's attention: this was the 
necessary result of circumstances, and does not deserve the 
blame which Livy attaches to it. The first law enacted, that 
henceforth one of the censors should necessarily be a plebeian ; 
this had in fact existed even before, for we know that C. Mar- 
cius was the first plebeian censor ; but it now became law, and 
was always observed : the second enacted that bills which were 
to be brought before the centuries should previously be sanc- 
tioned by the patricians whatever decree the centuries might 
think it right to pass. Formerly the consuls had the initiation 
in legislation; afterwards the praetor also had the same pri- 
vilege, since he likewise might preside in the senate and 
make proposals, his power being an emanation from that of the 
consuls; but the aedilcs, though they had the sella curulis, did 
not yet possess this right. A resolution passed by the senate 
on the proposal of a magistrate was not yet law, but had to be 
brought before the centuries and then before the curies ; this 
mode of proceeding arose at the time when the comitia of the 
centuries were instituted. The senate was formerly a patrician 
committee, and even now, the majority was undoubtedly patri- 
cian, though the plebeian element was already very strong. 
One hundred and ten years had elapsed since the decemvirate, 
and dviriug that period many patrician houses must have become 
extinct, and others must have passed over to the plebes. From 
Von Stetten's history of the noble families of Augsburg we 
see, that out of fifty-one families, thirty-eight became extinct 
in the course of 100 years, and that even then the surviving 
families made the same claims, which a hundred years before 
the llfty-onc families had been unable to establish. There was 
accordingly no reason for leaving to the patricians of Kome 
the veto which they had had before; and its abolition saved a 
great many unnecessary disputes. The more the patricians 
became reduced in numbers, and the more the ground tottered 
under their feet, the greater was their jealousy and the ill hu- 
mour which they introduced into the most important affairs of 
the state. The change made by Publilius, therefore, was very 


reasonable and necessary. But nothing was ever formally 
abolished at Rome ; when old institutions were no longer found 
useful, they were allowed to continue to exist as forms which 
could do no harm. Hence it was now enacted, that whenever 
the senate was going to pass a decree, the curies should sanction 
it beforehand. It is probable that this sham sanction was given, 
as in later times, by the lictors who were employed to represent 
the curies. The third law was, ut plebiscita omnes Quirites 
tenerent, and as I have explained before, affected such resolu- 
tions of the government {^rj^icr^aTa) as were to be sanctioned 
by the tribes instead of by the centuries. This, too, was a 
mere matter of form, for whenever the tribunes, who had 
previously consulted the consuls, were agreed among themselves, 
the plebes never refused their sanction. 

The following year, a.u. 417, was decisive, the army of 
Pedum with its neighbours and the inhabitants of the sea-coast 
being completely defeated by L. Furius Camillus and C. j\Iae- 
nius, and Pedum being taken by storm, C. Maenius is described 
by the ancients as the man who brought about the decision of 
the war: he gained a victory on the river Astura, the site of 
which is unknown ; a place of that name was situated between 
Circeii and Antium. It is quite certain that Maenius con- 
quered the enemies on the sea-coast, and Camillus those in the 
interior; and an equestrian statue was erected to the former as 
the conqueror of the Latin people. Henceforth no Latin army 
appeared in the field, and each of the towns capitulated for 
itself Livy's account of this seems to be extremely satisfactory, 
and the difficulties involved in it escaped me for many years ; 
but if we compare it with other authentic statements, it is by 
no means really satisfactory; for he assigns some events to too 
early a time, others are passed over, others again are described 
very vaguely, and lastly he makes no distinction between the 
free and the dependent municipium. Hence our knowledge of 
the relations of the Latin towns to Home is very imperfect. 
The whole of the Latin state was broken up ; but the lloman 
senate determined to preserve the separate towns and render 
them subservient to the interests of Rome: a plan which was 
carried out in different ways but witli extraordinary wisdom. 
Tusculum had from early times been in the enjoyment of the 
Roman franchise, though not in its full extent, but now its 
inhabitants received the full franchise; which was conferred 


upon the inhabitants of Lanuvium and Nomentum hkewise, 
who thus became full citizens like the Tusculans, their popula- 
tion being entered in the census lists as plebeians, and admitted 
into the tribes : the Tusculans were incorporated with the tribus 
Pupinia!^, the Lanuvians and probably the Velitemians were 
formed into a new tribe, apparently the Scaptia; whether the 
people of Nomentum constituted the tribus Maecia is uncertain . 
The Ariclnes, too, are mentioned by Livy among those who 
received the franchise; but according to an authentic account, 
they were, even some years later, in the condition of a depen- 
dent municipium. In this manner the places above mentioned 
attained great honours, and no town produced so many illus- 
trious plebeian families as Tusculum, though it was quite a 
small place; I need only mention the Fulvii, Porcii, Corunca- 
nii, Curii, and others'': certain places are particularly celebrated 
for the number of great men they have produced. At 
Lanuvium there was scarcely more than one family that 
acquired any celebrity. 

Other Latins likewise became citizens but not optimo jure, 
and this is the beginning of the class of citizens sine suffragio, 
which afterwards greatly increased and acquired a peculiar 
importance. The isopolites of the ancient times were niuni- 
cipes, and when they settled at Rome, they might exercise all 
the rights of Roman citizens, their position being similar to 
that of the citizens of the territory of Florence, previously to 
the year 1530. Those places which had received the civitas 
sine suffraffio, now stepped into this relation of isopolity. There 
was this dillerence, that formerly those only were municipcs, 
who came to Rome, but whose native place was perfectly 
independent in its political relations with neighbouring com- 
munities: this now ceased, and the separate towns which 
became municipia were perfectly dependent in all their foreign 
relations, whence Festus in his definition makes them form the 
second class of municipia. Such municipia had theconnubium 
with Rome and their own magistrates ; their inhabitants might 
acquire landed property in the territory of Rome, but were 
quite dependent upon Rome, like a son adopted by an-orjntio, 
or a woman quae in monum convenerat ; and in their relations 
with others they had no persona. Their rights in regard to 

' Also Popinia; sec Fcstup, .v. v. rujiiiiia tribus p. 233 ed. Miillcr. — Eo. 
' This is a remark of Cicero. — N. 


Rome were rights of conscience on the part of the Romans; 
they might acquire the Roman franchise by being personally 
admitted by the censors, but not being contained in the tribes, 
they did not serve in the legions : they were however obliged 
to furnish troops, not as allies (socii), but as Romani in separate 
cohorts. We may now ask whether they were required to pay 
the tributum, that is whether in case of the levying a tributuni 
being decreed at Rome, they had to pay according to the 
Roman census; and whether they had the right of sharing 
with the Roman people burthens and advantages, or whether 
their census was taken in their native places; the latter is 
probable, because they furnished and paid their own troops, and 
because the tributum was connected with the tribes. There 
cannot of course be any doubt as to their obligation to contri- 
bute. They unquestionably had a share in the public land, 
and whenever the Romans received a general assignment, those 
places too had a district assigned to them, with which they 
might do as they pleased. In this manner only can we con- 
ceive how Capua, after the war of Pyrrhus, could acquire such 
an extensive possession. 

The decision of the fate of Latium was an important epoch 
to the Roman state, for it gave rise to an entirely new class of 
municipia. The consequence was that many Romans purchased 
estates in those districts; but an inconvenience soon arose, 
inasmuch as these Romans had to submit to courts of justice 
composed of people who ranked much lower than themselves. 
This was afterwards remedied by the institution of a praefcc- 
tura, which the ancients, and especially Livy, misinterpret, as 
if the pracfectures had kept those towns in a complete state of 
dependence, whereas their real object was to administer justice 
to those who were full citizens of Rome. Such places were 
called fura or conciliabula, which is the same as what is called 
in America a town-house in any particular township: they 
were both markets and places for the administration of justice. 
A Roman, for example, who bought a slave at Capua according 
to Capuan law, had no right to call the slave his own at Rome; 
but when the purchase had been made in the presence of the 
praefect and according to Roman law it was unassailable. 

The fate of the other Latin towns was very severe. The 
ancient senators of" Velitrac, probably of "\'olscian descent, 
were led into exile beyond the Tiber, together with a large 


number of their fellow citizens, and a new colony was sent to 
Velitrae. A port colony was established at Antium; its 
inhabitants received the inferior Eoman franchise, and the 
Eoman colonists by settling there entered into the same relation. 
The Antiatans were deprived of their armed ships {interdictum 
mari), for the Eomans hated piracy; and this was the easiest 
way of getting rid of it, it being indifferent to the Romans 
whether the commerce of the Antiatans suffered or not. 
Among the remaining places, the connubium and commercium, 
as well as the common diets {concilia)^ were forbidden, just as 
in Achaia, Phocis, and Boeotia. No person belonging to one 
place was allowed to purchase land in another; but each town 
had its fixed burthens, so that if in one of them, in conse- 
quence of any calamity the price of landed property fell, the 
distress was very great, for the people of that place could sell 
only among themselves or to Roman citizens, the commercium 
existing with the Romans alone. This was the cause of the 
decay of those places, for as Romans settled in them the dis- 
tress became greater and greater, so that some of them entirely 
perished. Praeneste and Tibur alone maintained themselves: 
they were agro multati, but in the time of Polybius they again 
appear in possession of the ancient ^'m5 municipii. From Livy's 
accoimt it might be inferred, that the ancient alliance with the 
Laurentines had been preserved; and it is very possible that 
the same was done in the case of Praeneste and Tibur, so that 
they would have retained the right of municipium although 
their domain land was taken from them. Both possessed large 
and fertile territories, and must have had great vital power and 
energy : Praeneste tried more than once to shake off the Roman 
yoke. The punishment of isolation was also inflicted on all 
those places which at the close of the fourth century were in 
alliance with Latium; it extended moreover to the Aecpiians, 
who had undoubtedly been members of the Latin confederacy. 
The concilia remained forbidden, for the feriae Lafinae, 
formerly the general diet, became a mere shadow, a conventus 
(7rav7]yvpi<i) for the celebration of the games. 

Henceforth the Romans applied this system Avherever they 
wanted to break a conquered people, as they afterwards did in 
Achaia. The towns thereby became entirely separate ; the 
feeling of unity died away, they looked upon each other as 
strangers, and such a separation is usually followed by hostile 


feelings, as in Southern and Northern Dithmarsch. The Romans 
were obliged to have recourse to this Machiavellian system, as 
they placed no garrisons in the towns. It was in this manner, 
that the grand duke Peter Leopold of Tuscany, who kept no 
troops, separated his subjects and thereby demoralised them. 

The Latin colonies, it appears, were separated from the rest 
of Latium, whereas they formerly had been more closely con- 
nected with Latium, and were not in any direct relation w^ltli 
Rome; they now became a peculiar class of subjects, which had 
not hitherto existed at all. From this time forward Rome 
founded Latin colonies on her own account, and they deserve 
the admiration with which Machiavelli speaks of them, for they 
were the invention of great political tact. They were increased 
to the number of thirty, just as there had formerly been thirty 
Latin to^vns. The origin of these colonies was in the contract 
between the two nations: a district conquered by both in 
common used to be divided between them ; but districts which 
could not be thus divided, were set apart for colonies. Rome 
indeed foujided several colonies of her own; which received 
Caerite rights, but the former were called Latin colonies: 
Romans might settle in them, but they thereby stepped out of 
their tribes, though they might re-enter them whenever they 
pleased. Afterwards these colonies joined the Latin towns, 
and the thirty Latin places mentioned by Dionysius before the 
battle of the lake Regillus, were unquestionably the places 
named in the treaty of peace between Rome and Latium ; some 
of them were those towns which are said to have been founded 
by Tarquinius Superbus as Latin colonies, and which occur as 
such in the Hannibalian war. Now there can be no question 
that the Romans who had thus joined the Latins, obtained the 
equal franchise. The number of citizens in the Latin colonies 
was much greater than in the Roman ones. At a later time 
the Italians were admitted to a share in these colonics, and 
they sometimes obtained a portion of the domain land, so that 
the colonies became the great means of spreading the Roman 
dominion; and the Latin language, being the political language 
of the Romans, suppressed and supplanted that of the ancient 
inhabitants. They were from the first dependent upon Rome, 
and without any bond of union among themselves. Until 
the downfall of Latium the number of Latin colonies was 
insignificant, but from that time they began to increase. The 


inhabitants of all these places were bound to serve in the Roman 
armies, and Rome prescribed to them what numbers they had 
to furnish ; they were one of the principal means of the success 
of the Romans in the wars against the Samnites, for the Romans 
surroimded themselves with these colonies as with frontier 
fortresses. Several thousand men had a district assigned to 
them with the obligation of maintaining it; any Roman who 
wished to go out as a colonist, might do so, and others were 
added from Latium and other districts. The laws to be ob- 
served were prescribed by the Romans : the ancient inhabitants 
remained as a commonalty and undoubtedly formed the 
majority of tradespeople, but in a comparatively short time, 
they became amalgamated with the colonists, and these germs 
grew into a lofty tree. At first Rome etablished such colonies 
on the Liris in Campania, they were then extended into 
Umbria, and continued to be pushed onwards. This two-fold 
manner of founding colonies and conferring the franchise, 
sometimes with and sometimes without the suffrage, was the 
means whereby Rome, from being a city, became a state com- 
prising all Italy. The colonists paid no personal taxes, which 
devolved entirely upon strangers, they only paid the tax of 
the ager ex formula. 

The revolution which arose out of the conquest of the Latins 
was immense in regard to its consequences: even two years 
before, the destruction of Rome by the Latins was not an 
impossibility, but now her power was strengthened by those 
resources of Latium which had not perished in the struggle : 
but for the reasons already mentioned, the period which now 
followed, was for most of the Latin towns, a period of decay. 

Among the Campanians, likewise, the Romans produced 
divisions : they distinguished the Campanian populus (the 
equites who received compensation) from the plebes. The re- 
lation in whicli they stood to the Hcrnicans was not altered, or 
if it was altered, the latter had received a compensation in mo- 
ney in the victories of the Romans. Capua, Cumae, Suessulla, 
Atella, Fundi and Formiae became free municipia, that is, 
isopolite towns, and the Romans accordingly recognised, at 
least nominally, their perfect equality. 



Our accounts do not enable U3 to form a clear idea of the in- 
ternal condition of Rome : the war had cost her such heavy 
sacrifices, that, though her dominion extended from Sutrium 
and Nepet as far as Campania, the bleeding and exhaustion still 
continued for a long time: this renders the tranquillity which 
now followed quite intelligible, for all felt the want of peace. 

In the year after the decisive victory over the Latins (a.U.418) 
the praetoi'ship was divided between patricians and plebeians, 
on condition that certain forms should be observed, and from 
this time forward the praetorship, generally speaking, alterna 
ted between patricians and plebeians. This can be historically 
demonstrated : deviations from the law do indeed occur, but 
only serve to explain the rule. Q. Publilius Philo was the first 
plebeian praetor, and there may perhaps have been some con- 
nection between this law and the three which bear his name. 
When the second praetorship, commonly called the praetura 
peregrina, was added, one was always held by a patrician, and 
the other by a plebeian, just as afterwards when the number of 
praetors was increased to four, two were taken from each order. 
But when their number was raised to six, the equal division 
could no longer be kept vip, because the number of the patri- 
cians was ever decreasing. This law was the completion of the 
legislation of Licinius, for now the two orders were really placed 
on a footing of equality : great was tlie progress which had 
thus been made; for the fact that the patricians still continued 
to choose the intetreges exclusively from among themselves was 
a matter of no consequence. The repetition of the interregna 
at this time shows indeed, that the patricians still indulged in 
dreams of evading the law, for the charms of what they wished 
to gain increased as the number of those who laid claim to it 
diminished ; but these attempts do not appear to have called 
forth any violent reaction : the power of circumstances and 
truth were irresistible. 

Abroad Rome had no important wars to carr3''on; a trifling- 
one which broke out at this time was welcome to tliem, its 
object being to complete the compactness of their state as far 
as the Liris and Campania. The two banks of the Liris were 


inhabited by Auruncans (the Greeks call them Ausonians, and 
so also docs Livy when he follows Greek authorities, such as 
Fabius or Dionysius), an Oscan people. During the Latin war 
they had sided with the enemies of Rome, but afterwards they 
had submitted as subjects, and now were under the protection 
of Rome. The conquest of the Sidicines had been left by the 
Romans to the Samnites, but an arrangement seems to have 
been entered into, by which the Samnites allowed the Sidicines 
to continue to exist, in order not to lose the barrier between 
themselves and the Romans. This created a jealousy between 
the Romans and Samnites, and it could not in fact be other- 
wise: [the Samnite conquests in those districts had been the 
main cause why the Volsciang attached themselves to the Latins 
and afterwards to the Romans ; for at that time the Samnites 
were more dangeroiis to them than the Romans. Napoleon 
once said in a diplomatic discussion : '■^ il faut de petits etatsentre 
les (/rands" and on the same principle the large states allowed 
the small ones to make war upon one another, because this 
might lead to events calling for their powerful interference. 
These small states were, so to speak, '■'■pour les coups d'epingles 
qui precedent les coups de canons." The Sidicines united with 
the Auruncans of Gales attacked the other Auruncans, and this 
led the Romans to march against them. The latter carried on 
the war with great prudence ; they conducted it with lukc- 
warmncss, for it was anything but their interest to press the 
Sidicines, lest they should throw themselves into the arms of 
the Samnites. They took Gales, which is situated between 
Teanura and Gasilinum, and established a strong colony in the 
place. Their system now was to establish themselves by means 
of such colonies in the country between the Liris and Vultur- 
nus, so far as it was not already occupied by the Samnites ; and 
this system they followed out with great zeal and success : the 
colony of Gales connected the ever suspected Gampania with 
the dominion of Rome herself. A second colony was founded 
soon after at Fregcllae, which became so remarkable in the 
seventh century for its pride and its misfortunes; it was situa- 
ted on the spot where the Liris is crossed by the Latin road 
which leads through Tusculum to the towns of the Hernicans, 
and thence by Teanum to Gapua. The establishment of this 
colony was a true usurpation : the Samnites were masters of 
the country as far as Monte Gasino, they had there subdued 


the Volscians and destroyed Fregellae ; by their treaty with 
Kome they were permitted to make conquests in those districts, 
and even on their abandonincr them the Romans had no rio-ht 
to take possession of them. The Samnites had also taken Sora, 
and they had undoubtedly established themselves there with 
intentions just as ambitious as those of Eome. The Romans 
concluded a treaty of isopolity with the Caudines,and yet both 
nations were convinced that a war between them was unavoid- 
able. Under these circumstances, the Romans unquestionably 
adopted the same fluctuating policy which renders the history 
of the sixteenth century so interesting; the truth being: ^Hl y 
a trois sortes d'amis, ceux qui nous aiment, des amis indifferens, 
et des amis qui nous detestentT 

It is certainly not a mere accident, when we observe in his- 
tory that at certain times similar changes take place in countries 
far distant from each other : these changes in the one which 
produce a new state of things, cannot be the result of the changes 
in the other, because they occur simidtaneously and in different 
countries; we recognise in them the hand of Providence which 
guides the fate of men and the development of all nations as 
one great whole. The destruction of the Latin confederacy 
and the extension of the power of Rome is an epoch of that 
kind, and is quite similar to the period about the close of the 
fifteenth century. It is of great interest to compare the two 
periods : it is as if the stages of development through which 
particular countries and nations can pass by themselves had 
come to an end, and as if all their circumstances were to be 
changed by new relations ; for on such occasions we find na- 
tions joining one another which had never before been in con- 
tact, and states which till then had been most prosperous, begin 
to decay as if the autumn of their existence had set in ; the 
spirit of the most eminent nations becomes extinct never to re- 
turn : a change manifests itself in inclinations and tastes, and 
in the whole of the ordinary and daily habits of life ; nay even 
the physical nature of man undergoes alterations, for new forms 
of disease make their appearance. Such was the case about the 
end of the fifteenth century, for then the prosperity of the 
Itahan cities disappeared, just as at the time of which we arc 
now speaking, the states of Greece fell into decay. The very 
things which had been the cause of the prosperity of Greece, 
the equilibrium of the many small states, became the cause of 


her decay, no one single state being powerful enough to main- 
tain the independence of the whole. The circumstances of Italy 
in the fifteenth century were of precisely tlie same kind, for 
Florence and Venice stood by the side of each other with equal 
power, and if Venice had been strong enough to rule, a new 
and more beautiful order of things would have been the result. 
The battle of Chaeronea and the destruction of the Latin league 
occurred in the same year; and this simultaneously shows us 
the hand of Providence that rules the affairs of the world accor- 
ding to its own counsels. 

The Eomans and Samnites were apparently equal to each 
other, and there were reasons for believing that a struggle be- 
tween them would lead to the destruction of both, so that 
foreigners or barbarians would reap the advantages; for in the 
north the Gauls were already masters of a great part of Italy, 
and in the south the Carthaginians were threatening. Timo- 
leon, it is true, had a short time before checked the extension 
of their power in Sicily, but they were already masters of Sar- 
dinia with the exception of one mountain, and it was impossible 
to prevent them from acquiring sooner or later the islands of 
Sicily and Corsica. There was accordingly every appearance 
that after the mutual destruction of the Eomans and the 
Samnites, Italy would be divided between the Gauls and 

Until then no political relation had existed between the 
Greeks and Eomans ; but an intercourse with the inhabitants 
of Magna Graecia and the Siceliots seems to have been main- 
tained : I believe that even the literature of Magna Graecia 
exercised a much greater influence upon the Eomans than is 
commonly supposed, and at that time a knowledge of Greek 
was probably nothing extraordinary at Eome. Granting that 
Pythagoras did not become a Eoman citizen, since, perhaps, 
he is not even an historical personage, yet the Eomans were at 
an early period acquainted with the Pythagorean philosophy 
and entertained a veneration for it. Connections with the 
Greeks of neighbouring places are often mentioned ; Cumae 
aiforded ample opportunities, and the Sibylline books existed 
at Eome as well as at Cumae. The first embassies to the oracle 
of Delphi are fabulous, but there can be no doubt that the Eo- 
mans did consult that oracle. The connection of the Eomans 
with ]\Iassllia at the time of the Gallic conquest, and with the 


Lipariots, tlie guardians of the Tyrrhenian sea against the pi- 
rates, are the only other facts relating to the intercourse of the 
Greeks and Romans which we know for certain : all the rest is 
fabulous. But the first political affiiir in which the Romans as 
a state came in contact with the Greeks, belongs to this time ; 
for the treaty with Massilia was probably nothing but a com- 
mercial treaty, as I conclude, more particularly from the cir- 
cumstance that Massilia and Carthage were hostile to each other 
on account of the fisheries, as Justin relates ; by which we must 
understand either the coral fisheries on the coasts of Africa, or 
the tunny fisheries on the Italian shores. The inhabitants of 
Provence, throughout the middle ages, were in possession of 
the coral fisheries on the coast of Africa. The first political 
connection between the Romans and the Greeks was the treaty 
between Rome and Alexander, king of Epirus; for the Epirots 
may be regarded as Greeks, since notwithstanding their Pelas- 
gian origin they had become Hellenised. Alexander had been 
invited to come to Italy by the Tarentines in the year A.U. 420, 
or Olymp. 112. 

The glory of Magna Graecia had already disappeared; and 
most of the Greek towns, as Posidonia, Pyxus, Caidonia, Hip- 
ponium, Terina and others, had been conquered by the Luca- 
nians and Bruttians , some of them remained in the possession 
of the conquerors, others were abandoned : only a i'ew main- 
tained their independence, but had to fight for their existence. 
Rhegium, Locri, and the once flourishing Croton, had been 
laid waste by the Dionysii of Syracuse, who had abandoned 
those places indeed, but they were lying half in ruins and were 
but partially restored, as Delhi and Ispahan are at the yjresent 
time. Thurii and Metapontum defended themselves with dif- 
ficulty against the Lucanians; their territory was almost entirely 
lost, and they were struggling like the Italian towns in the sixth 
and seventh centuries against the Longobards. The only Greek 
town which, notwithstanding the general misfortune, was still 
in the enjoyment of the highest state of prosperity, was Taren- 
tum; this city too, soon after the period of the expedition of 
Xerxes, had suffered a great defeat from tlie neighbouring 
Messapians, but had soon recovered from it; and at the time 
when the tyrants of Syracuse and the Lucanians threatened 
the other towns, Tarentum was in a thriving condition; it was 
undoubtedly increasing by the immigration of numerous Greeks 

VOL. I. Z 


from the other towns, which were either destroyed or threatened. 
A parallel to this occurs in the growing prosperity of the Nether- 
lands and Switzerland in the time of the Thirty Years' War; 
the flourishing condition of those countries arose mainly from 
the distressing state of Germany, industry and commerce taking 
refiige in them. In this manner, Tarentum became wealthy 
and powerful ; it had, moreover, the additional advantages 
which a neutral state between two belligerent parties alw^ays 
has, and the rulers of the Tarentine state must have been men 
of great wisdom. 

The Tarentincs had acquired great wealth through their 
industry, commerce, wool-manufactures, their skill as dyers, and 
also from their salt works ; they had a powerful navy : and with 
the exception of Syracuse, no one of all the Greek cities, not 
even Rhodes, was as wealthy as Tarentum. Its inhabitants, 
according to their circumstances, were perfectly peaceable, and 
consisted of excellent seamen. There is no doubt that, as is the 
case with the people of the modern Taranto, navigation and 
fishing were their principal pursuits, a kind of idle busy life 
which is the delight of the Greeks and southern Italians: a 
Neapolitan is perfectly happy when he is rocking on the waves 
in his fishing boat. Nature has blessed the country about 
Tarentum with every thing in abundance. There is perhaps 
no part of the European seas so rich in fish and shell-fish as the 
bay of Tarentum ; and the poor Tarentine leads a truly princely 
life in idleness, for he requires only bread, salt, and olives, 
which he can always easily procure. The territory of Taren- 
tum was not so large as to lead the people to devote themselves 
much to agriculture. The tribes of the Latin race, the Etrus- 
cans, Umbrians, and Sabellians, on the other hand, were born 
husbandmen; and an Italian husbandman, who has an heredi- 
tary piece of land is thoroughly good, honest and respectable, 
while the people who live in towns are good for nothing. 
Those Italians who are not descended from Greeks are not at all 
fit for a seafaring life, and the Roman coasts were provided 
with fish by the southern towns, which continued to be Greek 
even in the middle ages. The Greeks arc bad husbandmen, 
and were so even in antiquity ; they cannot be compared to the 
Italians as agriculturists; the work of Theophrastus indeed 
shews great knowledge of agriculture, but the Greeks did not 
feel happy in that occupation ; they liked to cultivate the olive 


and vine but not corn. The soil of Greece, too, is in a great 
many parts almost unfit for the growth of corn, being better 
suited for the cultivation of olives. A Greek is cheerful and 
happy as a fisherman, and makes an excellent sailor. 

The Tarentines were quite a democratic people like the 
Athenians in Piraeus, as is observed by Aristotle ; and the state 
was very rich through the variety of its revenues. With these 
large means they were enabled to raise armies of mercenaries, 
as was then the custom throughout Greece, and as was the 
case in Holland in the seventeenth century. General opinion 
is not favorable towards the Tarentines: it is true that at the 
time when they became involved in war with the Komans, they 
were an eflfeminate and unwarlike people; but the censure 
which is usually thrown upon them arises from a peculiarity of 
human nature, which leads us, when a powerful state or indi- 
vidual falls, to seek for the cause of the fall in the conduct of 
the unfortunate, instead of feeling sympathy. I am convinced 
that next to Athens, Tarentum produced the wisest and most 
intellectual men in antiquity, and that the state made excellent 
use of them. A citj'- that produced an Archytas, the Leibnitz 
of his time, a man who possessed all knowledge then attainable, 
and was at the same time a crreat o-encral and statesman, — and 
neither envied nor banished him (as the Ephesians did with 
their Hermodorus) but raised him seven times to the office of 
supreme commander, shoidd not be censured: the spirit of 
Greece must have dwelt in it in all its beauty. The miserable 
anecdotes which Athenaeus for instance relates of the Taren- 
tines, are refuted by that one fact alone. They do not deserve 
blame any more than the great characters who are reviled in 
Schiller's Maria Stuart; a fact for which I cannot excuse 
Schiller, notwithstanding the beautiful poetry. It is certainly 
possible that Archytas and the other Tarentine statesmen 
looked more to the interests of their own city than to those of 
the Greeks in general (the Athenians alone rose to the moral 
height which enabled them to feel for all Greece); he may 
have kept up a good understanding with the tyrants of Syra- 
cuse, with more regard to the advantage than to the dignity 
of his native city; but these arc faults which the noblest men 
when placed at the helm of a state in unfortunate times have 
been unable to avoid. The Tarentines are blamed for having 
made use of foreign soldiers and armies, first of Archidauuis 

z 2 


of Sparta, next of Alexander of Epirus, then of Cleonymus, 
Agatlioclcs, and at last of Pyrrhus; whicli Strabo considers 
a sign of cowardice and imprudence; and lie at the same time 
adds the remark, that the Tarentines were ungrateful towards 
their protectors. But during the period that followed the Pe- 
loponnesian war, it was a general evil that wars were no longer 
carried on by armies of citizens, but by hired mercenaries : and 
this circumstance must be accounted for by the fact that wars 
had become more extensive and bloody, whereby the ancient 
race of citizens was destroyed. The ravages of war had made 
large numbers of men homeless, who wandered about, especially 
in Greece (as in modern times in Switzerland), by thousands, 
and were one of the greatest of plagues. It had long been 
a fair custom in Greece to leave the inhabitants of a town 
taken or destroyed in the enjoyment of their freedom, and not 
to sell them as slaves; but as all their property was taken from 
them, they were forced to gain their living in any w*ay 
they could : in the Thirty Years' War likewise, it became from 
year to year easier to find troops ^ ; such soldiers being con- 
stantly in arms were far superior to the militia, and when once 
they had begun to be employed, the militia soon became 
unable to resist the enemy. A city like Tarentum coidd not 
raise legions ; which can be formed only where there exists a 
respectable and numerous class of husbandmen, and hence 
there are countries where absolutely nothing can be done but 
to hire mercenaries, as was the case at Florence when the 
militia had got out of practice : but the same system would 
be destructive to other states. Tarentum therefore was under 
the necessity of making use of mercenaries^ and it would have 
been contrary to their notion of freedom to keep up a standing 
army; they acted wisely in confining themselves to their city 
militia when they could do without other troops. W^ienever 
there was a necessity for enlisting troops, numbers of homeless 
persons 2 were always to be got in Greece about Taenarus; 
they were however untrustworthy and faithless, since they 
followed him who paid best, like the condotti in the fifteenth 
and sixteenth centuries ; and a condottiere might easily act the 
part of a traitor or set himself up as tyrant. Hence it was 
much more prudent to engage the services of princes with 

Tl6\ffj.os vdKffxov rp(<p(i. ' Latrones, niado(p6pot 


their disciplined armies, for the honor of such a prince afforded 
at least some guarantee. Why should the Tarentines have 
disturbed their commerce and trade, as they were enabled to 
manage things differently? Such a hired army might indeed 
become dangerous, but so long as it was possible they took 
wise precautions: Alexander of Epirus afforded them real 
advantages, but Pyrrhus did not. The English system of 
levying armies has likewise been censured, but only by persons 
who had no knowledge of the circumstances of the country. 
There is nothing that deserves censure in the conduct of the 
Tarentines except the insolent manner in which they drew 
upon themselves the war with Rome, but we shall assuredly 
have no reason to be severe towards them, if we consider the 
exasperation which drove them to it. 

About this time they became involved in a war with the 
Lucanians, who had attacked Heraclea and Metapontum, which 
towns were under their protection. The Lucanians had 
already lost that part of Calabria, which was afterwards called 
Bruttium, for its inhabitants consisting of the Pelasgian serfs 
of the Greek towns had united into one people, and refused to 
obey the Lucanians, who were wise enough to recognise their 
independence and remain their friends. But in order to in- 
demnify themselves, the Lucanians turned their arms against 
Tarentum, attempting to subdue Heraclea. In these circum- 
stances the Tarentines invited to their assistance Archidamus 
of Sparta, who with the unhappy Phocian exiles had gone to 
Crete; but he fell in an engagement against the Lucanians, on 
the same day on which the battle of Chaeronea was lost. A 
few years afterwards they took into their pay Alexander the 
Molossian, of Epirus, a brother of Olympias, the wife of 
Philip. Philip had given him his daughter Cleopatra in 
marriage, and had allowed him an appanage: his kingdom was 
very small; Philip who everywhere contrived to gain strong 
positions, kept the fortress of Ambracia for himself, and at 
first gave to Alexander only three small towns in Cassopia, on 
the Thesprotian coast; afterwards, when Philip had extended 
his empire and every where put himself in possession of the 
fortified places, he raised Alexander to the throne of the 
Molossians, among whom he found but little to do. Philip 
followed the same policy in regard to his relations, as Napoleon 
did ill reference to his brothers: they were to be kings, but 


without power; so that tliey were notliing else than satraps 
without paying tribute. It was for this reason that Philip 
retained Ambracia for himself. During the time that Alex- 
ander of Macedonia was engaged in his Eastern expedition, 
Alexander the Molossian was under the authority of the inso- 
lent old Antipater; he was not on good terms with the 
]\Iacedonian king, and according to the accounts of the ancients, 
it was jealousy of the glory of his nephew, that induced him 
to go to Italy ; he is said to have complained bitterly that fate 
had made him fight against men, while his nephew was opposed 
only by women. As the Macedonian was not inclined to allow 
our Alexander to extend his dominion in Epirus, the latter 
received the invitation of the Tarentines with great pleasure. 
He accordingly went to Italy, but with intentions quite differ- 
ent from those with which the Tarentines had invited him : 
they expected that he, as a small prince, with a well trained 
army^ would protect them, but Alexander went over with a 
desire of conquering a kingdom for himself He was sviccessful ; 
subdued the Messapians and Salleutines, made a diversion to 
Posidonia, delivered the Greek towns, and united them into a 
confederacy, of which, he of course, became the aTparrj'yo'i 
and Tjje/jicov. Being in the service of the Tarentines, he was 
never in want of subsidies, like the nations who in the 
last century were in the service of England under Walpole; 
but the history of his exploits is almost entirely lost, and it is 
only in the Greek grammarians such as Tzetzes, that a few 
interesting statements are preserved. His success was brilliant, 
so long as he acted in concord with the Tarentines; but when 
he betrayed his ulterior intentions, and wished to assume the 
title of king of Italy (of course in the narrower sense of the 
term), the Tarentines were exasperated and dispensed with his 
services. Whether they concluded a separate peace with the 
Lucanians is uncertain ; but as the diet of the Greek towns 
met at Heraclea, although Tarentum was the most powerful 
and illustrious of those towns, the diet seems to have been 
transferred by Alexander, which clearly indicates a rupture 
between him and the Tarentines. However, as his power was 
now too small, he seems to have carried on the war as a mere 
adventurer like Charles XII. : he made predatory excursions, 
and Pandosia in the heart of Lucania became his Pultawa; 
there he was surrounded by the Lucanians and Bruttians: his 


army was divided, both parts were annihilated, and he himself 
was slain. He had previously concluded a treaty with the 
Eomans, which is incidentally mentioned by Livy, but uu- 
doubtedly on the authority of Eoman annals, and this treaty 
is a proof of the manner in which the Eomans made their 
calculations: they had nothing to fear from him, and sought 
his alliance only for^the purpose of overawing the Samnites, 
who had concluded a treaty with Tarentum. A real alliance, 
however, did not exist between Eome and Alexander, for the 
treaty between Eome and the Samnites was still in force. 
So far as we know the circumstances, we must blame the 
Eomans for having favoured a foreigner in preference to 
kindred people of their own peninsula. The Samnites indeed 
are not mentioned among those who in the end waged war 
against Alexander, but his predatory excursions had brought 
him into contact with them : at Posidonia they fought against 
each other. It is interesting to speculate on what would have 
been the probable consequences, if Alexander had established 
a kingdom in Italy; it is likely that he would only have 
facilitated the victories of the Eomans, and hence their treaty 
with him was very prudent, though not praiseworthy. 


The ancient historians had no difficulty in forming a clear 
conception of the relations then existing between Eome and 
Saranium, as we see especially in the excerpts de Legationibus 
from Dionysius : each nation saw in the measures of the other, 
nothing but fraud and hostility, and on the whole they may not 
have been very wrong in these suspicions. The Eomans had 
kept the nations that dwelt about Campania, partly in the 
condition of isopolites, such as the Fundanians and Formians, 
and partly in a state of dependence, as the Privernatans. These 
latter endeavoured to shake off the yoke; for the franchise 
without the suffrage was only a burthen for them, and. the 
advantages which they enjoyed were small in proportion to 


what they cost ; the light of acquiring landed property within 
the Roman dominion, was no benefit to a city which had itself 
a fertile territory. The Romans imagined that this insurrection 
had been stirred up by the Samnites; and there can be no 
doubt that every one dissatisfied with the government of Rome 
met with sympathy from the Samnites. The Privernatans 
were joined by the Fundanians, one of whose nobles, Vitruvius 
Vaccus, was the leader in this movement; but the Fundanians 
did not persevere, and withdrew from the contest. The 
Privernatans were severely judged by the Romans, of wliich 
a very interesting account is given by Livy and Valerius 
Maximus : the ambassadors of the Privernatans were asked to 
state conscientiously what punishment they had deserved ; they 
answered that they deserved the punishment due to those who 
struggled for liberty. The consuls received this answer favor- 
ably, and then asked whether they would keep peace if they 
were pardoned, whereupon they replied: "If you give us an 
honorable peace we will keep it, but if you give us a degrading 
one we shall break it." The consuls then said, that men like 
these deserved to be Roman citizens, and the franchise was 
accordingly conferred upon them. The same story occurs in 
the excerpts de Legationihus of Dionysius, but many years 
earlier, and there is perhaps no foundation at all for it. It is 
related by Valerius Maximus indeed, but he is no authority 
whatever, being only an echo of Livy. The story is perhaps 
an invention of the gens Aemilia or Plautia which had the 
patronage of Privernum, and bore the surname of Privernas, 
the annalists having afterwards inserted it where they thought 
fit.^ A few years afterwards, the Privernatans, according to 
an unequivocal expression in a plebiscitum", were again in a 
state of insurrection ; bi;t this has been eifaced from history in 
order to preserve the interest of the old story. At a later time, 
Privernum was in possession of the franchise, and that of a 
higher kind than the mere Caerite franchise, for they formed 
the tribus Ufentina. Fundi and Formiae too were severely 
punished. This is the natural connection of the events so 
pathetically narrated by Livy : the generosity which he ascribes 

' The Plant! i pix'scrvcil upon their coins tlic recollection of the conquest of 
Privernum as the most glorious eveut in tlie histoiy of their family. Rom. 
Hist. Ill, p. 175. L. Aemilius Mamorcimis Privemas, and C. Plautius Decianus 
triiimphcii OA-er the Privernatans. — En. ' Liv. vni. 37. 


to the senate is quite incredible, and his account of it a piece 
of mere declamation. 

There is no doubt that the Samnites secretly promoted the 
disturbances among the subjects of Rome, and they openly 
demanded the evacuation of Fregellae. Justice was un- 
questionably on their side, for the Romans had no right to 
establish a colony in a place which had been conquered by the 
Samnites, although at the time when Rome sent her colony 
thither, it was not in the hands of the Samnites ; for otherwise 
they would, perhaps, after all not liave sent it. But in such 
cases justice cannot always be done : wrong and injustice are 
often very different things. On this occasion, I should not like 
to cast a stone at the leaders of the Romans for not giving up 
a place which they had taken in a deserted district, even if 
their taking it was an act of positive injustice. The Samnites 
were rapidly spreading in that district; and Fregellae, at the 
head of a bridge on the upper Liris, was a strong point for 
defending the country against them; and the advantage which 
the Romans might derive from its possession was much less 
than the disadvantage to them of Fregellae being in the hands 
of the Samnites. As soon as Rome gave it up, the Latin road 
would have been opened, and her allies, the Hernicans, Latins, 
and undoubtedly the Aequians also, would have been exposed 
to imminent danger. The case was similar to that of 1803, 
after the peace of Amiens, when the evacuation of Malta by the 
English was demanded by everybody : the English could not 
give it up, though they had promised it, which surely they 
ought not to have done. The slow movements of the Samnite 
senate might, perhaps, have been some security against any 
abuse that might be made of Fregellae. 

The outbreak of the war was so anxiously looked forward to, 
that even two years before it took place, a Roman army was 
encamped on the frontier, it being expected that the Samnites 
would make an attack upon Fregellae. By the treaty with 
Alexander of Epirus the Romans hoped to secure a friend, and 
they now tried to protect themselves against the enemy, still 
more by a peace with the Gauls. The latter had now been 
settled in Italy for upwards of sixty years, the migrations 
across the Alps became every year less numerous, the commotion 
among those nations had ceased , and the Ga;ils who were never 
an entirely savage people did not fail to acquire a certain 


civilisation; they devoted themselves to agriculture and became 
a harmless agricultural race, just like the Goths under Vitigis, 
who were likewise a defenceless host of peasants, whoni Totilas 
was obliged to prepare for war by special training. The Gauls 
had before them two roads to the South of Italy, — the marshes 
about the Arno and the wild part of the Apennines protected 
Etruria; — the one down the Tiber through Umbria towards 
Latium and Campania, and the other through Picenum along 
the coast of the Adriatic towards Apulia. Upon this latter 
road they must have repeatedly attempted to advance southward : 
but it was more difficult than the other road, because they were 
opposed by the Sabellians in the Abruzzo, Now the Komans, 
in order not be distiu-bed by an attack of the Gauls, which the 
Samnites might easily have brought about, concluded a formal 
peace with them, which Livy passes over in silence, but which 
is expressly mentioned by Polybius, and which the Romans 
undoubtedly purchased with money, for why should the Gauls 
have promised the Romans to remain at peace? 

This anxiety of the Romans to protect themselves for the 
future, renders it highly probable to me that the old statement 
of their having, in conjunction with the other Italian nations, 
sent an embassy to Alexander at Babylon, is not a mere fiction. 
Alexander had put a limit to his conquests in the East, and to 
march Southward against the Ethiopian nations would have 
been senseless. It was naturally expected, when he returned 
from India that he would direct his arms against the West; 
for no one was so foolish as to believe that he would all at once 
put a stop to his conquests. Many persons are of opinion, that 
the people in the West knew nothing at all of the Eastern 
conquests of Alexander ; but the Western nations were not so 
much isolated from the rest of the world as is generally thought : 
the Romans must have known of the expedition of Alexander, 
just as Clapper ton and Dcnham found people in the interior of 
Soudan, acquainted with the insurrection of Greece and with 
the co-operation of individual Europeans in it. During the 
Seven Years' War, when my father was in Sana, people there 
had a very distinct knowledge of the great war that was being 
carried on in Europe, and especially of the war between the 
English and French ; nay, one intelligent Arab of Sana brought 
forward a map, and made enquiries about the geographical 
position of those European states. This happened in the very 


heart of Arabia ; and we roust remember tbat the modem Arabs 
are a degraded and ignorant people, which cannot be said of 
the nations of antiquity. The means of communication, more- 
over, were far more easy at that time than in the twelfth and 
thirteenth centuries, when, nevertheless, there existed communi- 
cations with the remotest parts of Asia. I believe that the 
Komans had accurate information about Persia and India: it is 
true they did not yet possess geographical works, but they 
undoubtedly had maps of the world like those which existed 
in Greece, and it is certain that at that time some Eomans 
received a Greek education, as seems to be proved by the very 
surname of P. Sempronius Sophus. No one doubts that the 
Samnites and Lucanians sent ambassadors to Alexander, 
although later writers call in question the statement that the 
Eomans did go ; the Lucanians sent their embassy in order to 
avert his wrath on account of the death of his uncle; the 
Samnites, in order to secure his friendship if he should come 
to Italy; and in like manner the Komans were anxious not to 
offend him at least, though they might not hope to win his 
friendship. Even the Iberians sent an embassy to him as soon 
as they heard of his preparations against Carthage. Livy hits 
upon the singular idea that the Eomans had perhaps never 
heard of him. It is possible that the Eomans concealed the 
embassy from pride, or that the Greeks invented it from vanity ; 
but it would be necessary to suppose that the latter was done 
at a time when the Eomans were already so powerful that the 
homage of Eome could increase the glory of Alexander. But 
Clitarchus, through whom the account of the embassy has come 
down to us, was an elegant author ; he wrote immediately after 
the death of Alexander, at the time when the Eomans were 
still engaged in the doubtful contest with the Samnites. 
Aristobulus and Ptolemaeus Lagi, who far surpass him in 
historical fidelity, speak of Tyrrhenian and Samnite ambas- 
sadors, and the former of these names comprises the Eomans 
also, just as the name of the Samnites applies to all the Sabellian 
nations. If Alexander's life had been spared, he would have first 
directed his arms against Sicily, and thence against Carthage, 
which would certainly have fallen before him: in Italy the 
Greeks would have received him with the same enthusiasm as 
in Asia Minor, for he was Setvo9 TrapeXKetv; he would have 
won them, concluded treaties with them, and have weakened 


those wlio opposed him so much, that the whole peninsula 
would have been his. Livy has a discussion upon this point 
which is very beautifully written, but a complete failure: his 
national vanity entirely blinds him, and he is egregiously 
mistaken in his calculation of the military resources of Italy, 
as well as in his belief that all Italy would have united against 
Alexander. If he had come to Italy, Rome would certainly 
have fallen ; and his death was a necessary ordinance of Provi- 
dence in order that Rome might become great. 

This was the state of affairs at the time when the war broke 
out between Rome and Samnium. The immediate occasion of 
the war was the conduct of Neapolis and Palaepolis, the ancient 
Parthenope. Palaepolis is mentioned only by Livy: it was an 
ancient Cumaean colony, the Cumaeans having taken refuge 
there across the sea. Neapolis derives its name from being a 
much later settlement of different Greek tribes, and was perhaps 
not founded till Olymp. 91, about the time of the Athenian 
expedition to Sicily, and as a fortress of the Greeks against the 
Sabellians. It is not impossible that the Athenians also may 
have had a share in it. Both towns, however, were of Chalcidian 
origin and formed one united state, which at that time may 
have been in possession of Ischia. Many absurdities have been 
written about the site of Palaepolis, and most of all by Italian 
antiquaries. We have no data to go upon except the two 
statements in Livy, that Palaepolis was situated by the side of 
Neapolis, and that the Romans had pitched their camp batween 
the two towns. The ancient Neapolis was undoubtedly situated 
in the centre of the modern city of Naples above the church of 
Sta. Rosa; the coast is now considerably advanced. People have 
sought for Palaepolis likewise within the compass of the modern 
city, without asking themselves whether there would have been 
room for an army to encamp between the two places. I alone 
should never have discovered its true site, but my friend, the 
Count de Scrre, a French statesman, who in his early life had 
been in the army and had thus acquired a quick and certain 
military eye, discovered it in a walk which I took with him. 
The town was situated on the outer side of Mount Posilipo, 
where the quarantine now is; it is an excellent and healthy 
situation facing the islands of Nisida and Limon : it may be 
that in antiquity there was a port at Palaepolis, and the two 
islands still have very good harbours. That point moreover 


had a natural communication with Ischia. Mount Posilipo 
with its ramifications lay between the towns at a distance of 
about two miles, so that there was sufficient room for the 
Koman army to encamp on the hills and thus to cut off the 
communication between the two towns. There exist neither 
monuments nor coins of Palaepolis. According to the common 
supposition, the two towns would have been so near to each 
other, that darts thrown from their walls would have met in 
their course. 


The war was occasioned by piracy or at least by hostihties 
committed at sea against the defenceless merchant vessels of 
the Romans, who had then no fleet, and strangely enough pre- 
tended not to care for the sea, as if such things could be 
neglected with impunity. Complaints respecting the division 
of the Falernian territory may likewise have contributed to 
the outbreak of the war. Such a division was always a great 
event; many persons sold their lots, while others settled upon 
their farms; and that district became a sore point in the do- 
minion of Rome. If, however, this was the cause of the dis- 
ruption between the Palaepolitans and Romans, the complaint 
of piracy which Dionysius introduces in so declamatory a man- 
ner is entirely out of place, since it would be no more than 
natural for them to endeavour to disturb the commerce of the 
hostile people. The Neapolitans, trusting to their alliance 
with the Samnites and Nolanians, refused the reparation which 
the Romans demanded of them. The Oscan population had 
gradually become predominant at Xola, though it had undoubt- 
edly Chalcidean epoeci, who formed a considerable part of the 
population. How much its inhabitants had become Hcllenised 
may be seen from the Greek symbols on the coins of Nola with 
the inscription NflAAIflN. It is in general remarkable, how 
easily the Samnites amalgamated with the Greeks ; Strabo calls 
them (f)LX,i\\r]ve<;, and the Samnites, without a literature of 
their own, were undoubtedly open to that of the Greeks, and 


endeavoured to speak like the Greeks themselves. The Ko- 
mans never agreed well -with the Greeks, to whom the Luca- 
nians also were hostile although their civilisation was Greek; 
and it certainly cannot be questioned that the Pythagorean 
philosophy was established among th(im. The statement that 
Pythagoras was a native of one of the Tyrrhenian islands must 
mean, that the roots of the theological parts of his philosophy 
must principally be sought for among the Pelasgians and in 
the religion of Samothrace. 

An auxiliary corps of 4000 Samnites and 2000 Nolanians 
threw themselves into the towns of Palaepolis and Neapolis; 
the Tarentines are likewise said to have stirred up Palaepolis, 
for the Tarentines who were very well disposed towards the Sam- 
nites employed their money to involve the Romans in war at 
a distance. The Romans looked upon the occupation of Palae- 
polis by the Samnites as an act of hostility, and brought their 
complaints before the diet of Samnium. The evacuation of 
the place was a moral impossibility, and the answer which the 
Samnites returned was, that as the Romans wished for war, 
war they should have, and there was no need to dispute about 
trifles. This answer was confirmed by the assembly of the 
Samnite people. In the meanwhile the siege of Palaepolis 
had already been going on for some time, and the Romans had 
no prospect of success, for their art of besieging was still in its 
first infancy, and the Greeks opposed them with great technical 
skill; the attacks of the Romans therefore produced no eflfect, 
and the sea was left open to the Greeks. But treason did what 
force was unable to accomplish. NeapoHs possessed ships of 
war with which they may frequently have made predatory ex- 
cursions against the Roman coasts, which the Romans were 
unable to protect. The Samnite garrison, at least the greater 
part of it, appears to have been stationed at Palaepolis, and the 
Greeks at Neapolis. Two Greeks Charilaus and Nympheus 
now betrayed the Samnites to the Roman consul Publilius 
Philo : they proposed to the Samnites to make an expedition 
against the Roman coast, and the Samnites quitted the city 
ready to embark. As the town on the side of the harbour was 
protected by a wall, the conspirators closed the gate after the 
Samnites had gone out, and admitted the Romans by another 
gate. The Samnites found that the ships had been drawn 
away from the coast, and were obliged to save themselves as 


well as tliey could. Palaepolis now disappears from history, 
and tlicre can be no doubt that it was destroyed on that occa- 
sion. Neapolis obtained a favourable alliance with Rome, 
from which we may perhaps infer that the conspirators were 
Neapolitans. The acquisition of Neapolis was extremely im- 
portant to Eome; for thus the two harbours of Nisida and 
Naples, from which alone expeditions by sea could be under- 
taken against their territory fell into their hands. At that 
time Naples was not, as at present, a city of 400,000 inhabitants, 
but must have been somewhat like our town of Bonn. This 
conquest was made by Q. Publilius Philo pro consule ; he was 
the first to whom the consular power was prolonged (a. U. 42 9) 
by a senatus-consultum and a plebiscitum, on the proposal of a 
tribune, so that his own law concerning plebiscita was applied 
to him. The fact of a new magistracy being thus created in 
substance though not in form, was a great change in the con- 
stitution. Up to that time no one had celebrated a triumph 
except during the period of his magistracy, but Publilius 
triumphed as proconsul. 

This was the beginning of the second Samnite war which, 
if we except the Hannibalian, is the greatest, most attractive 
and most noble in all the history of antiquity. It is to be 
lamented that we know so little about it, for the places where 
the battles were fought are mostly passed over in silence ; but 
we should not be ungrateful; with some pains we may yet 
obtain a satisfactory knowledge of the war. Livy had described 
some parts of it with great pleasure, but others with evident 
weariness, which was the result of his mode of writing: he 
entered on his task without preparation, whence he wrote with 
freshness and vigour indeed, but had neither a clear insight 
into the histoi'y nor a command of his subject. If he had 
made better use of the annals, we should see our way more 
clearly. It is to be regretted that the books of Dionysius on 
this war are lost, for the few fragments in Appian, who copied 
from him, and in Constantine Porphyrogenitus, throw inuch 
light upon many points; Dionysius' account of that period 
must have been excellent, for the annals were already sufficient 
to enable a diligent searcher like him to make out a real his- 
tory. There existed some nameless chronicles as early as that 
time, though they may have been dry and obscure in their 
details : the fact that isochronistic history does not commence 


till a hundred years later, is here of no consequence. Livy 
has unfortunately made no use at all of the ancient materials 
which formed the foundation of the annals whence he makes 
his choice quite arbitrarily when the annals contradicted one 
another, and in most cases prefers that which is wrong. He af- 
fords us no means of gaining a general view of this war which 
lasted twenty-two years; and it was only after many years' 
study that I succeeded in forming a clear conception of it. 

The war must be divided into several periods : the first ex- 
tends from the year A.U. 429 to 433. During this period the 
Samnites appear to us in a strange light ; for although they had 
wished for the war, yet they were evidently unprepared, and 
seem to have had the conviction that they would not be able 
to hold out. The instigators of the war must have lost their 
popularity, and the war itself was disagreeable and troublesome 
to the people. Such a state of things may appear surprising; 
but those who have witnessed the great war of the revolution 
must remember quite similar circumstances. The case of the 
Athenians in the Peloponnesian war also resembles that of the 
Samnites, for after the first and second campaigns they wished 
for peace; and so did the Venetians after the battle of Ghiera 
d'Adda. In the year 1793, the war against France was quite 
popular in England (I was myself in England about that time), 
for the English remembered the interference of France in the 
American war, and still had great expectations fi:om their 
colonics : the national hatred too was generally speaking very 
great, although a few were in favour of the revolution ; but 
when the war was carried on badly, when no objects, at least 
no important ones, were gained, and when the power of France 
was ever increasing, the war became thoroughly unpopular, 
and the general outcry was for peace, so that in order to main- 
tain themselves, the ministers were obliged to yield and enter 
into negotiations. When, however, the nation became aware 
that peace was impossible, they rose in a brilliant manner, and 
in 1798 and 1799, the war was again extremely popular. This 
observation is very humiliating to those who attach so much 
importance to public opinion. Such also was the case with 
the Samnites, for when it was proved that the Eomans carried 
on the war quite differently from what had been expected, the 
Samnites were disappointed and wished for peace. Afterwards, 
however, a complete change took place in their minds, for as 


the war was protracted, they began to feel as if they couhi 
not live without it, especially when it was carried on unsuccess- 
fully, for, as in gambling, men will rather perish than withdi'aw 
from a contest, and thus give themselves up to the enemy; and 
this feeling changed the war into one of guerillas. In the midst of 
the war, when these misfortunes were much greater than at 
the beginning, the Samnites had arrived at the conviction that 
peace was impossible. 

The Samnites, as has already been remarked, consisted of 
four states, which took the supreme command in turn. This 
was a very great disadvantage, for when one general was 
elected, the other leaders probably hated and envied him, for 
such is always the case among allied states, as we see in the 
history of the German empire and the United States: may 
God prevent this ever happening in the army of om- German 
confederacy ! The unhappy war of the revolution likewise 
arose from the fact, that in the campaign of 1799 one general 
rejoiced in the defeat of the other. When a great man like 
Pontius had the command, and it so happened that the other 
praetors were honest men and acted with him, a great advan- 
tage might indeed be gained ; but in the year following every 
thing was altered again. If the Samnites had been tmited, 
they would have been more than a match for the Romans ; 
but as it was, the Romans overcame them through the 
excellence of their institutions, for various and even most 
hostile elements were all firmly concentrated under the one 
power of the spirit of Rome. In the art of war, the Samnites 
undoubtedly equalled the Romans, for, according to Sallust, 
the Romans had adopted their armoiu" and perhaps their whole 
mode of warfare from the Samnites, at least we find that in 
the battles the annies were drawn up in exactly the same 
manner, and the reports of the battles attest that they 
fought against each other as equals against equals. I must 
here contradict the opinion of General V^andoucourt, who 
asserts that the Italian, Spanish and African nations fought 
their battles drawn up in the phalanx. Their strength consist- 
ed in the sword : the Italicans had cohorts, and undoubtedly 
used the pila like the Romans. 

The Samnites it appears, had allies; the district from 
Frentenum to Lucei'ia being either an allied country or a distinct 
canton; but the alliance was so loose that the Frentenians kept 

VOL. I. A A 


entirely aloof during the war. To the north of the Sainnites 
there existed the confederacy of the Marsians, Marrucinians, 
Vestinians and Pelignians; and of these the Vestinians were 
on friendly terms with the Samnitcs, while the others were 
indifferent to them and even attached to the Romans. The 
situation of the Samnites was thus very perilous, but if they 
had carried on the Avar on the Liris as far as Capua, which 
seems in fact to have been their plan, they would have been 
able to have maintained themselves against the Romans; but 
the latter had a far bolder plan : for, as in the Latin war, they 
again formed a semicircle round Samnium, a plan which now 
involved much more danger and was on a greater scale than 
in the Latin war. The Samnites were cordially hated by the 
Apulians, among whom the ruling class consisted of Oscans, 
who may either have subdued the ancient Pelasgian population 
and amalgamated with them, or have expelled them. The 
country of Apulia is surrounded by mountains which form a 
horse-shoe, so that the country presents the aspect of a theatre ; 
the mountains themselves form a part of Apulia, but the real 
country below these mountains is a table land of a chalky soil, 
and almost as hot as Leon in Spain. The Apulians had two prin- 
cipal towns, Arpi and Canusium, each of which was the mistress 
of a large territory and jealous of the other. The Samnites had 
conquered the eastern hills of Luceria; and the plain, too, 
may have been threatened by them. As Tarentum was allied 
with the Samnites, the Apulians applied to the Romans, and 
much may have been gained by their mediation. It was a 
gigantic resolution of the Romans to transfer their army to 
Apulia: thei'e were two roads, the one passing through the 
country of the Aequians, who were friends of the Romans, 
along the lake of Celano, by Sulmona and through the narrow 
country of Samnium ; the other led through the country of 
the Sabines to Reate, Civita Ducale, and the fearful passes of 
Antrodoco (the ancient Interocrea) which are of such a kind 
that a gallant people may there resist an enemy for a very long 
time, but which were so disgracefully abandoned by the Nea- 
politans in 1821; the road then proceeds to Pescara on the 
eastern coast, and thus reaches Apulia by an enormous circuit. 
The two roads were probably taken by the Romans at different 
times, but at first they marched along the former; now as long 
as they were not sure of the Vestinians, but were on good 


terms with the other nations, they certainly could take the 
former road; for on the latter, the Vestinians were the only 
one of the four northern Sabellian tribes through whose country 
they were obliged to pass in order to reach Apidia; in addi- 
tion to which they would have had to fight their way through 
the territory of the Frentenians. But if they had chosen the 
former road, the Marsians and Pelignians would, unquestion- 
ably, have opposed them as much as the Vestinians, since it 
was their interest not to alloAV the Romans to march into 
Apulia. Xow, as on that occasion the Vestinians are called 
peaceful, it is clear that the Romans marched through the 
passes of Antrodoco. Had the Samnites been united, they 
ought to have made every effort to support the Vestinians ; but 
this was not done, in consequence of which the Romans 
defeated and compelled them to submit. They therefore estab- 
lished themselves in Apulia, and thereby obliged the northern 
confederates to keep up a good understanding with one another. 
It was a great advantage to be in possession of Apulia : the 
country of the four Sabellian people as well as that of the 
northern Samnites, the Pentrians, Bovianians and even of the 
Frentenians, is a mountainous and pasture country in the 
Abruzzo. During the winter those districts are covered with 
snow, and it is impossible to keep sheep there ; whence during 
the winter they are sent into Apulia, which is then covered 
with beautiful and excellent grass: in the spring the shepherd 
drives his flock again into the mountains. In southern 
countries the great features of nature always remain the same, 
and they are at the present day just what they were in antiquity. 
The establishments at Tarentum for dyeing wools show that 
the breeding of sheep was very extensive as early as that time. 
The use of those pastures was of the highest importance to the 
Marsians, Marrucinians, Pelignians, etc., and the Romans, being 
in possession of Apulia and protecting the pastures for their 
allies, obliged them to maintain a friendly understanding, and 
at the same time pressed hard upon the northern Samnites. 
Hence we see that the Romans did not undertake that formid- 
able expedition at random, but that their course was thoroughly 
justified by the nature of the country; nevertheless, they did 
not venture upon the hazardous undertaking, until they saw 
that it was unavoidable ; and that this was the only Avay in which 
the war could be brought to a close. 




The Romans had formed an alliance with the Lucanians as 
well as with the Apulians. The Lucanians are called a Sara- 
nlte colony, which must be understood in a different sense 
from what we mean by a colonial city. It is certain that the 
Lucanians were an offshoot from the Samnites, from whom 
they had separated themselves. They dwelt among the 
Oenotrians (the ancient Pelasgians) and Greeks; and as the 
Samnites were Sabellians who had become Oscans, so the 
Lucanians were Oenotrians who had become Samnites: they 
had commenced extending themselves about Olymp. 80, that 
is, at the time of the fall of Sybaris, which opened those 
districts to the Italian nations. We have no information 
respecting the relation in which the Lucanians at first stood 
to the Sanmites. The territory of Lucania is larger than that 
of Samnium; but there was not a corresponding proportion in 
the powers of the two nations, as we see from the census lists. 
The Lucanians were never powerful, not even in much later 
times, when the Samnites were greatly reduced; the number 
of their capita did not amount to 30,000, that is much less 
than half that of the Samnites. This shews that the greatest 
part of the Lucanian population had no share in the sovereignty, 
which was concentrated in a few places only, such as Petelia; 
the country was distracted by parties. One portion of the 
people resolved to join the Romans; but this can have been 
only a small majority; for soon afterwards a revolution took 
place in which that alliance was broken, and the Sam- 
nites were invited to occupy their fortified places. We are 
acquainted with the treaty of the Lucanians and Romans from 
Livy, but all the rest of his narrative relative to these events 
must be greatly modified, as for example when he says, that 
the Tarentines, frightened by the power of the Romans, pre- 
vailed upon the Lucanian nobles to tell the people that the 
Lucanian ambassadors had been cruelly treated by the Ronians, 
at which the people are said to have been enraged, and actuated 
by this feeling, to have thrown themselves into the arms of 
the Samnites. This is the same story as is related of Zopyrus 
and Sextus Tarquinius. We here see that treacherous blind- 


ness of party spirit, which is so saddening in the history of the 
later Greek states. The Samnites thus unexpectedly became 
masters of Lucania, and availed themselves of its resources, 
both in men and money, for their own advantage. 

These wars, as far as we can survey them, are from the 
beginning extremely intei'esting, on account of the determina- 
tion, skill, and firmness with which they were conducted. 
They resemble a single combat between two excellent cham- 
pions, for the two parties aimed at each other's life, directing 
their blow with the greatest boldness at each other's heart. 
They fought with the same resolution as in modern times has 
been shewn in attacks upon particular places. If after the 
battle of Cannae, Hannibal, with his enormous talent, had had 
the same resolution — if he had not been too cautious, but had 
followed the same plan against Rome as the Samnites did, he 
would decidedly have triumphed over his enemies. Each of 
the belligerent parties calculated very much upon the dis- 
affection of those who were dependent upon the other. The 
frontier of the Samnites was in the Abruzzo above Sora, and 
Casinum was their city. The course of their operations seems 
always to have been determined by those mountains. Thence 
also they acted on the offensive, and that with the definite 
object of causing an insurrection among the Latins, who 
fourteen years before had been independent, and y^eve there- 
fore inclined to rebel. The traces of a partial insurrection are 
obscured in Livy, but are nevertheless discernible ; and we 
find, that even Tusculum, in conjunction with Privernum and 
Velitrae, rose in arms; but the Romans always quelled these 
insurrections, and the consequence was the destruction of 
many of the Latin towns. All this can be inferred also from 
certain allusions; for example, from the proposal of a tribune 
(which, however, was not carried), to destroy the Tusculans 
altogether. To the same circumstances must be referred the 
strange story in Livy of a sudden nocturnal alarm in the city, 
as if the enemy were within the walls; for as the armies were 
at a great distance, an insurrection of the Latins naturally 
produced terror up to the very walls of Rome. 

The Samnites endeavoured to penetrate through the Apen- 
nines to the sources of the Liris, and straightway to advance 
towards Rome. The Romans at the same time crossed the 
Vulturnus, and tried to reach Saticida in Campania, and 


tKence to invade Samnium. Each was little concerned where 
the blows of the other fell, provided it could itself inflict a 
deep wound. This method of carrying on the war had pecu- 
liar advantages for both. For the liomans it was an advantage 
that the Samnites rav.-iged only the territory of their allies, 
whereas the Komans inflicted sufferings on the Samnites 
themselves. This, however, could not have the same evil 
effect as the ravages of the Samnites produced upon the dispo- 
sition of the Roman allies. It is a mere accident that we 
know that the seat of this war was in the neighbourhood of 
the modern abbey of Subiaco, on the frontier of the Aequians 
and Hernicans, among the high mountains which separate the 
valleys of the Liris and the Anio. Livy states^ that the 
enemies faced each other near Imbrinium, in Samnium; but 
even the early Italian commentators, such as Sigonius and 
Hermolaus Barbarus, justly remark that Imbriviimi must be 
meant; and they identify the place with that from which the 
emperor Claudius constructed his aqueducts, in the country of 
the Aequians, near Subiaco. Livy shews too few traces of 
accuracy and care to prevent us adopting this correction, 
which is commended not only by probability, but by positive 
necessity. There the Samnites established themselves, and 
thus cut off the Romans from the road to Apulia, whereby 
the latter were obliged to keep up the communication on the 
road by Antrodoco. That district is very important in mili- 
tary history. Circumstances wore so dangerous an aspect, 
that in the third year of the war, the Romans appointed 
L. Papirius Cursor to the dictatorship, the consul L. Furius 
Camillus being ill. Papirius Cursor is remembered among 
the first generals of his nation. By his side stood M. Valerius 
Corvus, who was of about the same age, and the younger 
Q. Fabius Maximus, whom Valerius Corvus probably survived. 

M. Valerius Corvus was the most popular man of his age. 
He was free from all political party spirit; he loved the 
people, and was beloved by them, and the soldiers had 
unlimited confidence in him. In his leisure hours he felt as 
happy among his soldiers as in the midst of his family; he 
shared his labours and his pleasures with them : his popularity 
was the inheritance of the Valerii. It was his personal 
character that enabled him to quell the insurrection of the 
year A.U. 413, which no one else woidd have been able to 


L. Papirius Cursor was a rough, and properly speaking a 
barbarous man, who had somewhat of the character of Suwarow, 
except that the latter was a far more educated man, for he was 
well acquainted with German, French, and English literature, 
and possessed great judgment. Cursor had enormous bodily 
strength; and, like the emperor Maximinus, kept it up by 
eating and drinking like an athlete. He tormented and 
annoyed the soldiers by excessive severity, and rendered their 
service as hard and diflicult as possible, thinking that the 
soldiers would thereby become all the more useful. Towards 
the officers and commanders of the allies he was equally 
severe. It was his delight to see those around him tremble; 
and he would not pardon the slightest neglect, but inflicted 
corporal and even capital punishment upon those who were 
guilty of it. He was generally hated, and looked upon as a 
demon, in whom, however, the republic possessed an invin- 
cible bulwark, which in case of need might afford a last 

Q. Fabius was a different man from Valerius Corvus. He 
does not appear to have been so cheerful and loveable a 
character as Valerius; but he was withal comis, a gentle com- 
mander, and a mild and wise man. Great reliance was placed 
on his wisdom and good fortune; in the latter, Papirius was 
inferior to him. He too was highly popular, but not in the 
same manner as M. Valerius ; for it seems to have been owing 
more to the respect than to the love which was felt for him. 
He was regarded as the first man of his age, whence he 
received the surname Maximus. fie was no less great as a 
statesman than as a general, and was a rallying point for all 
parties. By birth and rank he was an aristocrat, but a very 
sensible one, and in many cases he was able as arbitrator to 
bend the oligarchy. His whole life shews that he was in 
earnest in everything, and able to control his own feelings, 
and sacrifice them to the good of the commonwealth. It is 
these tliree men who give to the history of that period its 
peculiar interest. 

The dictator faced the Samnites in the neighbourhood of 
Subiaco, but at the same time another army was stationed in 
the neighbourhood of Capua to protect Campania against the 
inroads of the enemy. The dictator is said to have perceived 
that the auspices had not been correctly observed. It was im- 


possible for him to take new ones where he was, the auspices 
differing according to the localities, some being valid at Eome, 
others in the enemy's country; hence he was obliged to return 
to Rome to take fresh auspices on the Capitol. Wliether it 
really was for this, or any other reason, he left the camp and 
went to Rome, leaving the command to Q. Fabius, his master 
of the horse, expressly enjoining him not to act on the offen- 
sive. This injunction may have been well founded ; but it is 
not impossible that it arose from a want of confidence in the 
}'Ounger man, or from a desire not to allow him an opportunity 
of distinguishing himself The Samnites very soon observed, 
that the Romans were not permitted to fight, and they there- 
fore provoked and pressed them all the more : the inactivity of 
the Romans was dangerous^ for the Latins in their rear were 
ever ready to revolt, if the Samnites should offer them support, 
though by themselves they were unable to undertake anything. 
Under these circumstances, Fabius with youthful confidence 
resolved to give battle to the Samnites : he gained the victory, 
and according to some authorities even defeated the enemy 
twice. As the army regarded the dictator's order only as the 
result of his ill-will and envy, the master of the horse sent his 
report, not to Papirius Cursor, but direct to the senate, disre- 
garding him who had the auspices, and through whom alone 
the report ought to have been sent. He then burnt the booty, 
in order to deprive the dictator of the spoils for his triumph. 
In the city the fear of the consequences was undoubtedly not 
less than the joy at the victory. Papirius forthwith returned 
to the camp; and his speedy arrival there shews that the army 
cannot have been far away from Rome. Surrounded by his 
twenty-four lictors, he summoned the master of the horse before 
his tribunal, and only asked him whether he had fought against 
his orders or not. When every thing was ready for the exe- 
cution of Fabius, the whole army assumed so threatening an 
attitude, and the general iudignatiou at Papirius was so great, 
that he himself began to hesitate, and at the urgent request of 
the soldiers, granted a respite until the following day. In the 
night Fabius fled to Rome and applied to the senate ; but during 
its meeting, and while Fabius was standing in the midst of the 
hall, Papirius himself also appeared, and demanded his victim. 
Although the senate afterwards shewed on several occasions 
that it was not favourable to Fubius, yet sympathy for the 


youthful hero was then very general, and it was resolved to 
protect him. Papirius did not dare to use force: the situation 
of Fabius, however, was not so desperate as Livy describes it, 
for we know from Verrius Flaccus, that the patricians had the 
right of appeal from the verdict of the dictator to the curies. 
Livy's statement that he invoked the tribunes, is either a con- 
fusion caused by the expression, provocatio ad popuhim, or it 
was a sanction of the decree of the curies by the plebes, in 
which case the whole people would have granted an amnesty 
to Fabius. Papirius even now refused to yield, but the deter- 
mination of the two orders snatched his victim from him. 
Livy's statement, that he became reconciled to Fabius, is im- 
possible. Fabius resigned his magistracy, and Papirius took 
another master of the horse. The object of general hatred, 
he returned to the army, and the unfortunate issue of an en- 
gagement was attributed to him. This happened in the year 
A.u. 430. 

Fabius is said to have gained his victory chiefly by having 
ordered i\\Q frena to be taken from the horses, and thus caused 
the cavalry to dash upon the enemy. If by frena we under- 
stand reins, the statement would be absurd : and the difficulty 
may be explained by the bits which have been discovered at 
Herculaneum and Pompeii. The bridles and bits of the horses 
used by the Romans were extremely cruel ; if therefore instead 
of these, the Roman general ordered the gentler ones of the 
Greeks described by Xenophon to be used, it is natural that the 
horses, thus eased, should have pressed forward with greater 
cheerfulness and vigour. 

The war took such a tvirn that the Samnites were in great 
difficult}'-, and regretted having undertaken it. They concluded 
a truce on condition of their giving pay and clothing to the 
Roman soldiers, and then began to negociate for peace, which 
they thought they might obtain by yielding to the first de- 
mands of the Romans in reference to the garrison at Neapolis 
and the recognition of the colony at Fregellac. But the Ro- 
mans now undoubtedly made quite diffijrent claims, demanding, 
in addition, that Lucania and Apulia should be evacuated, and, 
what was always done in such a peace, that the Samnites 
should be reduced to the same position as if they had been en- 
tirely subdued: this was one of the maxims which contributed 
to the greatness of the Romans. The attempt to conclude a 


peace was unsuccessful, the war was renewed, and the Romans 
now conducted it with great energy. Fahius, who was made 
consul, led his army into Apulia and took Luceria and many 
other towns of the Apulians and Samnites. His repeated vic- 
tories compelled the Samnites to withdraw from Fregellae in 
order to oppose his progress. The other Roman army was also 
successful, and as the Romans gained great advantages in the 
whole of the following campaign, the Samnites came to the 
determination to seek peace at any cost. They now vented all 
their indignation upon Papius Brutulus, the man whom they 
regarded as the soul of the whole war, and who belonged to 
the family which, two hundred years later, produced C Papius 
Mutilus. The Romans asrain concluded a truce, for which the 
Samnites made great sacrifices. We are indebted to the ex- 
cerpts from Dionysius for a knowledge of these transactions: 
the Samnites were ready to do every thing in their power to 
punish the authors of the hostilities; but the Romans unques- 
tionably demanded the surrender of Papius Brutulus, and the 
resolution which he took shews that he was a great man. He 
had lived for his countrymen, and served them as long as they 
wished to be great, but now that they were desponding, life 
had no value in his estimation, and he made away with himself, 
in order that his fellow-citizens might be able to say that the 
author of the war had atoned for his offence. This is one of 
the most heroic acts in all antiquity, and is greater even than 
the similar deed of Cato. The Samnites, to their own disgrace, 
sent his body to Rome. 

As the Romans had, on the first application for peace gone 
beyond the demands they had made before the war, so they 
now again exceeded the terms they had last proposed, demand- 
ing that the Samnites should recognise the supremacy of Rome 
{inajestatem populi Romaiii comiter colere). The Samnite 
ambassadors had appealed to the humanity of the Romans, 
they had declared tliat they would accept any terms, if the 
Romans were resolved not to give up a single point, but that 
they could not consent to recognise the Roman supremacy, 
since upon this point the national diet alone could decide. 
The consequence of such a recognition would have been a 
state of perfect dependence in all their relations with foreign 
states : they would have been obliged to give up their alliance 
with the Tarentines and Lucanians; and Roman commissioners 


would have appeared among them with the right to enquire 
whether the treaty was duly observed. Sucli terms were 
intolerable to the Samnite people : they had lost their leader, 
humbled themselves, and imploringly prayed for a suitable 
peace ; but all was now in vain ; they resolved to perish to a 
man rather than conclude such a peace as was offered to them. 
This time the Komans had carried their maxim too far: the 
Samnites exerted their utmost power, and commenced the war 
in Apulia on account of the physical importance of that coun- 
try. Luceria with its Roman garrison was besieged: it had 
originally been a Samnite town, but had been conquered by 
the Apulians. The Romans also changed their mode of warfare, 
and as the main army of the Samnites was stationed in Apulia, 
resolved to concentrate all their forces too in that country: 
they had before directed their attention to Ap^ilia, and had 
indeed found some allies there, but without gaining a firm 
footing. They would accordingly have been obliged to compel 
the Vestinians to allow them to march through their territory, 
a plan which seemed to be dangerous, because they might thus 
become involved in a war with the Marsians, ]\Iarrucinians, 
and Pelignians. But there the unfortunate jealousy among 
the Samnite tribes would have come to their assistance; other 
nations also to whom the Romans were troublesome, such as 
the Aequians and even the Campanians, sided in their hearts 
with the Samnites, though they did not wish the latter to gain 
a decided victory: those little petty nations imagined that the 
Romans and Samnites would mutually weaken each other, and 
that they themselves might derive advantages from this state 
of things. 

When it was known that Luceria was besieged, both consu- 
lar armies wanted to march to Apulia, and resolved to take 
the nearest way, forcing their road through the midst of 
Samnium, for the Samnites had become contemptible in their 
eyes. They perhaps took the road by which A. Cornelius 
Cossus had gone, viz. the one from Capua by Beneventum to 
Luceria. C. Pontius, the general of the Samnites and one of 
the greatest men of antiquity, who had foreseen this, left at 
Luceria only as many of his troops as were necessary to con- 
tinue the blockade, and encamped on the road Avhlch the 
Romans had to pass, near Caudium, the capital of the Caudine 
Samnites: that town afterwards disappeared from the face of 


the earthy tliat there might be no trace of the disgrace of the 
Romans. Tlic Romans descended a defile into a valley, on 
the opposite side of which another pass formed a steep ascent 
up the mountain : they had not yet met the enemy anywhere, 
and therefore advanced very carelessly. The army, forming a 
long column, had descended the one pass, and the first part of 
the column was beginning to ascend the opposite defile, but 
found it completely barricaded with stones and trees. The 
Sajnnit^s had probably made preparations of the same kind as 
the Tyrolese in the year 1809, who had placed on the heights 
large trunks of trees fastened together with ropes, and behind 
them huge blocks of stone, so that when they cut the ropes 
the enemy in the valley below were cmshed under the falling 
masses : this seems to be suggested by the mention of stones 
in Livy. According to his account, the Romans behaved on 
this occasion in a most cowardly ma]iner, for they are said to 
have attempted to return, and finding that the opposite path 
was likewise obstructed, they made up their minds to encamp 
in the valley. This is an absurdity, for an army thus shut in 
would under all circumstances fight with the courage of despair 
and endeavour to escape. There can be no doubt that a 
pitched battle was fought in which the Romans were defeated, 
as is clearly stated by Cicero {cuvi male j^ugnatum ad Caudiwn 
esset). Appian, of whose work we have only fragments rela- 
tive to those events, states that the superior oflicers who 
survived with the consuls, signed the peace; he mentions 
twelve tribunes, but as the complete army contained twenty- 
four tribunes, twelve must have fallen, or at least have been 
severely wounded. Zonaras, also, speaks of a lost battle and 
the conquest of the Roman camp. In urging the point that 
there was no engagement at Caudium, Livy displays a truly 
strange kind of vanity : he describes the Romans as cowards 
in order to conceal the disgrace of a defeat. The particulars 
of this affair are buried in great obscurity, but the result of 
my investigations is as follows. According to Livy's account 
the consuls only promised that the Roman people woidd con- 
clude peace, and that beyond this nothing was agreed to; so 
that he represents the Romans as not having been faithless; 
but that half of the Roman equites (six hundred) were given 
as hostages. But the afiair was in reality quite dififerent: 
Appian, who derived his information from Dionysius, says that 


tKe hostages were given , eto? a7ra<i 6 Brifj,o<; Tyv eipr'jvrjv i7n-\jn]- 
(f)i(rrj, that is, until the curies and tribes should have ratified 
the peace. Its terms were fair; C. Pontius, not knowing, in 
the extreme joy of success, what use to make of it, summoned 
his father Herennius Pontius, a friend of the Tarentines and 
especially of Arcliytas* into his camp, to ask him how he 
should treat the Romans. Herennius answered that all should 
be cut to pieces; and when the son replied, that this was inhu- 
man, the father is reported to have advised his son to dismiss 
them all without injury, in order to place the Romans under 
an obligation by this act of grace. But the Romans of that 
time would have laughed at such an evtjOeia. The meaning 
of the story can only be this: — Herennius meant to say "The 
only thing that can be done, is to destroy the enemy; how 
can you have any doubt about that? If you are at all in 
doubt, you had better dismiss them at once." But C. Pontius 
was a high-minded man, he had a great Italian feeling, and it 
was impossible for him to annihilate the army of a nation 
which protected Italy against invading foreigners, especially 
Gauls and Carthaginians; he did not doubt that a lasting 
peace might be concluded with the Romans, if they could 
be secured; we fortunately knoAv its terms from the frag- 
ments. The consuls and all the commanders pledged their 
word of honour that the people would ratify the peace ; and 
imtil then the equites, the sons of the most distinguished fami- 
lies, were to remain as hostages, the status quo ante bellum was 
to be restored, all conquered places were to be given back to 
the Samnites, the colonists were of coui'se to be withdrawn 
from Fregellae, and the ancient equal alliance between the Ro- 
mans and Samnites was to be renewed. Compensations in 
money or any humiliating conditions are not mentioned at all ; 
the Romans were to depart, but leave behind all their arms, 

' Herennius appears to have been altogether a model of wisdom among the 
Samnites. According to a passage ia Cicero, de Senectute, he was one of tlie 
interlocutors with Archytas in a philosophical dialogue of some Pj'thagorcan 
philosopher, — a rem;xi-kable proof to what extent tliose Italiote towns were 
familiar with the Sabellian people, and how little they looked upon tliem as 
barbarians. For the Opicans they had a great contempt, and probably made a 
marked distinction between them and the Samnites. Tlie intercourse with the 
Greeks explains liow it came to pass, that Nimia, the source of all Subdlian 
wisdom, was regarded as a Pythagorean: this is a genuine Sabine tradition. 
They went so far ia their friendly feeling, that tlie Greeks insisted upon the 
Samnites being a Spartan colony. — N. 


money, waggons, horses, etc. This is in accordance with 
the general Italian law of nations. The passing of the 
Romans under the yoke is described as superbia on the part 
of the Samnites, but was quite in the natural course of things : 
the Samnites had completely surrounded the Romans with 
palisades: some of these were taken out, and a gate was formed, 
through which the Romans were allowed to pass one by one 
unarmed. The same thing had often been done before, and 
was perfectly natural. It should be remarked, that Pontius 
was so far from being cruel, that, according to Appian, he 
granted to those who departed, sumpter horses to carry the 
wounded to Rome and provisions for their journey. Never 
has a great victory been more nobly used. The question now 
is whether the peace was ratified by the Roman people, for 
here lies the cause of so grave a charge, that Livy places it in 
the back ground. The fact of the peace having been ratified 
is attested by the circumstance that the tribunes of the people 
were delivered up to the Samnites; they accordingly must 
either have sanctioned the decree of the curies regardino- the 
peace, or have made a formal proposal to the plebes for that 
purpose. A tribune of the people was not allowed to spend a 
night out of the city ; and therefore could not have been among 
those who had concluded the peace with tlie enemy in the 
camp. The only other possible way of explaining the circum- 
stance would be to suppose, that by a formal decree, a tribune 
was sent to the army; but even this can be conceived only on 
the supposition that he was sent thither for the purpose of 
ratifying the peace. This was necessary in order to recover 
the hostages, and therefore the peace was ratified, to be after- 
wards broken, under the pretext that the consul and tribunes 
who had brought the motion before the senate and the plebes, 
were traitors and ought to be delivered up to the Samnites. This 
is the most detestable act in Roman history, and surely the 
Romans had good reason to conceal it; in order to do this, 
Livy has corrupted and distorted the history of the whole of 
the year following, by stating that in it the Romans, at the 
conquest of Lucretia, recovered their hostages, who considering 
that the peace had been so shamefully broken, would certainly 
have been massacred long before. 



The existence of the peace is further attested by the events 
which took place afterwards; for in the very next year we (hid 
the Samnites in possession of Luceria and Fregellae : it is said, 
indeed, that the latter place was conquered, but this may be a 
forgery, or the colonists w^ere unAvilling to quit their homes, 
and the Romans may then have left the place to be taken by 
the Samnites. At any rate the latter occupied Fregellae, which 
was a matter of great importance, if the war should break out 
again; for Fregellae commands the Latin road leading from 
Tusculum through the country of the Hernicans to the upper 
Liris and Campania. The Romans therefore now had only the 
road by Terracina, Lautulae and the lower Liris in the 
neighbourhood of Minturnae : moreover when a Roman army 
was stationed in Campania, and another marched by Subiaco 
into Apulia, the communication between the two was cut off. 
Of still more importance was the subsequent occupation of Sora 
by the Samnites, not only for the reasons already mentioned, 
but because they thereby acquired a basis for their operations. 
The calamity of Caudium belongs to the year of the city 433, 
according to Cato ; and this forms the conclusion of the first 
period of the war.^ 

The Romans now cancelled the peace, and delivered up to 
the Samnites the consuls and other commanders who had sworn 
to it : by this means they endeavoured to escape the punishment 
for their perjury, and it was perhaps for this purpose that they 
had carried their hypocrisy so far as to cause the peace to be 
decreed by the tribes and not by the centuries, in order to 
exclude the auspices, and thus to avoid coming into collision 
with the law of religion. Livy, on the occasion of the surrender 
of the tribunes, indulges in a perfectly senseless piece of 
declamation : the tribunes had to meet their fate as well as the 
consuls, and in so deep a humiliation of their people, they 
could hardly look upon their personal misfortunes as anything 
extraordinary. It is further related, that the consul Postumius 
kicked the fetialis who delivered him up to the Caudines, with 

' In the Lectures of 1826-7, Nicbuhr fixed the end of the first i^eriod hefore 
the defeat of Caudium, so that the second period would be that of the success 
of the Samnite arms. -Ed. 


these words: "Now the Romans may carry on the war with 
justice, for I am a Samnite citizen and liave violated the law 
of nations." This sounds quite absurd, but it is nevertheless 
possible, for we know from Velleius Paterculus, that previously 
to the outbreak of the war isopolity had been established with 
a portion of the Samnites, and these Samnites may have been 
those very Caudines; now as every Roman on going into exile 
mifjht assume the franchise of such a state, Postumius, accordino- 
to the forms of the law of nations, may have claimed for himself 
the franchise of the Caudines. By such a detestable farce he 
imagined that he was drawing the punishment of heaven upon 
the Samnites. But however this may be, the peace was broken 
in a most unprincipled manner, and this act forms a glaring 
contrast with the noble generosity of C. Pontius, who sent back 
all the prisoners, saying, that if this principle was to be followed, 
the Romans ought to send all their legions back to Caudium, 
in order that the affair might be restored in integrum, and that 
the individuals were not his enemies. This shews Caius Pontius 
to have been an extraordinary man, and the Samnite people to 
have possessed great moral worth. 

The Samnites continued to gain great advantages, but none 
that were lasting, and the Romans, who made immense efforts, 
returned to their former plan of operation, that is, they conducted 
the war against Samnium from Apulia and on the western 
frontier. Publilius Philo and L. Papirius Cursor were elected 
consuls : the latter went to ApuUa ; the former is said to have 
fought on the road which Avas so unfortunate for the Romans 
in the year 433, and to have forced his way to Papirius who 
was stationed near Arpi. This is not very probable, but we 
cannot speak with any certainty about it. The Romans 
established themselves at Arpi which was friendly to them, and 
from it they carried on the siege of Luceria. There Pontius 
is said to have been blockaded with 7000 Samnites and the 
600 Roman hostages, to have been obliged to capitulate, and 
to have been dismissed after having passed under the yoke. 
But the whole story is nothing but an invention of vanity. 

Diodorus' accounts of these times deserve great attention; we 
know not whence he derived his materials, it may be from 
Fabius or from Timaeus; that he made use of the latter at 
least, is very possible, for Timaeus may have written the histoiy 
of this period as an introduction to his history of Pyrrhus, or 


in bis histories of Sicily and Italy. The statements of Diodorus 
are very remarkable though they are extremely fragmentary 
and unequal. He sometimes drops the thread of his narrative 
and takes it up again at random ; be is on the whole a very 
miserable historian; bis work contains names of places which 
are now quite lost : some are evidently mistakes and perhaps of 
the author himself, but others are simply unknown to us. 
Livy's account of the year 434 (the consuls at that time entered 
upon their office in September, so that what he relates belongs 
to the spring of 435) occurs in Diodorus under the year 439, 
which is far more probable, for it is not likely that Luceria 
was conquered twice. The consuls undoubtedly confined 
themselves to making preparations, and reducing to obedience 
those of the allies who had become rebellious. The Eomans 
now made the greatest exertions in Apulia, most of whose 
inhabitants they subdued; for in A.u. 436 and 437 there 
was a truce between them and the Samnites, which had been 
effected by the mediation of the Tarentines, who were greatly 
concerned about the restoration of peace, since they dreaded 
lest the Romans should permanently establish themselves in 
their neighbourhood. The trvice at this period was a mis- 
fortune for the Samnites; and there can be no doubt that 
C. Pontius was not invested with the supreme command, 
owing to the jealousy of the other cantons. The Romans 
already began to assume an imposing attitude, but in A.U. 438, 
the war again burst forth with extreme violence. It is full of 
the most remarkable vicissitudes of fortune ; the ever memor- 
able campaign of the year 1757 indeed is more brilliant, but 
we anight also compare it with that of the Samnites. They 
conquered Sora by treachery, whence Ave see that, pursuing 
the same plan as they had adopted at the beginning of the 
war, they again endeavoured to extend their sway on the 
Upper Liris. The Romans, on the other hand, with that 
lionlike intrepidity which characterises both nations in this 
war, laid siege to Saticula in the neighbourhood of Capua, for 
the purpose of gaining ground against Samnium, and disturb- 
ing the Samnites by a diversion. I may here pass over the 
detail. One Roman army was already in the interior of 
Samnium, and the other in Apulia, but both were almost 
surrounded, so that a report of the danger reached Rome. 
The Samnites had strengthened themselves on the Liris, and 

VOL.1. B B 

370 Q. FABIUS. — Q. AULIUS. 

the Romans^ perceiving tliat it was tlie object of all their 
movements to cut off Campania from Rome, sent a detach- 
ment under the dictator Q. Fabius, with the greatest haste, to 
the pass of Lautulae, whence he was to join the array in 
Campania. But even Fabius was not invincible. The Sam- 
nites came across the mountains behind Fundi and occupied 
the narrow pass, the Thermopylae of that country. The 
Romans, who seem to have fallen in with them unexpectedly, 
were completely defeated and put to flight, as is clearly stated 
by Diodorus (a.u. 438 or 439); Q. Auhus, the master of the 
horse, allowed himself to be cut to pieces. This victory pro- 
duced a mighty revolution, for the Samnites now spread into 
Latium. Satricum joined them, and the nations, far and 
wide, either actually revolted, or showed a hostile disposition. 
In what manner fortune turned is a point on which Livy 
leaves us in the dark, because the preceding defeats are only 
slightly alluded to by him. The Samnites were besieging a 
place which Diodorus calls Kinna (we do not know what place 
is meant). The Romans, in relieving it, completely defeated 
the enemy, and then again subdued the revolted towns. One 
of the revolted people were the Ausonians or Auruncans, 
about the mouth of the Liris, who had probably intended 
to remain neutral. Some of those who may have been most 
compromised now displayed features of baseness which one 
would hardly think possible. Twelve Auruncans came and 
sun-endered their towns to the Romans, who destroyed them; 
which Livy, with his kindly feeling, relates with horror, but 
in a political point of view the destruction was quite right. 
The more difficult the circumstances were, the more necessary 
was it for them to strike terror into their subjects, for they 
could not calculate upon any attachment. Livy says Deleta 
Ausonum gens vix certo defectionis crimine, an expression we 
cannot perhaps take in its strict sense. The disposition to 
rebel extended as far as Praeneste, the revolt of which place in 
this very year may be inferred from Livy, for under A.u. 449, 
in speaking of the Praenestine Q. Anicius, who was then 
plebeian aedile, he says, qui panels annis ante hostis fuerat. 
But most of these people, in going thus far, only injured 
themselves without benefiting the Samnites. None of them 
wished that the sovereignty of Rome should pass into the 
hands of the Samnites, but all were anxious to remain separate 


between the two in their miserable independence. If they 
had been prudent they would have endeavoured to unite with 
Rome, and Rome would readily have received them. It is a 
pity that Livy passes over these painful reports, and does not 
explain in what manner the two Roman armies contrived to 
escape from their perilous situation. This must have taken 
place, and deprived the Samnites of their advantage. Livy 
himself says, omnes circa populi defecerant. We are indebted 
to Diodorus for our knowledge that the army of Fabius saved 
the Romans. By dint of a careful examination, we can in 
some measure determine the whole extent of the insurrection. 
According to Diodorus, Capua actually revolted ; while accord- 
ing to Livy it was only suspected, and the leaders of the 
conspiracy made away with themselves. The former state- 
ment is more probable ; and the consequence was, that a 
Roman army under C. Maenius, who was appointed ■praetor rei 
gerendae causa, marched into Campania and re-conquered the 


In the year 440, which is the turning point, the second 
period of the war came to its close. The battle of Lautulae 
and its consequences had raised the Samnites to the summit of 
prosperity; but the Romans now again succeeded in drawing 
fortune over to their side, as in general they always shewed 
themselves greatest after a misfortune. Horace says, merses 
profunda pulchrior evenit ; the Romans never lost their pre- 
sence of mind except after the battle on the Alia. With such 
a determination they could not fail to conquer the world. 
He who is at one with himself, able, conscious of his power, 
and who resolutely resists his opponent, is always sure to win. 
Even the very next year, Rome paralysed her enemie